Ensuring Value for Money in Health Care: The Role of Health Technology Assessment in the European Union (Who Regional Office for Europe) [1 ed.] 9289071834, 9789289071833, 9789289071840

Citation preview

European on Health Systems and Policies

ENSURING VALUE FOR MONEY IN HEALTH CARE

The role of health technology assessment in the European Union

Corinna Sorenson, Michael Drummond, Panos Kanavos

Observatory Studies Series No 11

Ensuring value for money in health care

The European Observatory on Health Systems and Policies supports and promotes evidencebased health policy-making through comprehensive and rigorous analysis of health systems in Europe. It brings together a wide range of policy-makers, academics and practitioners to analyse trends in health reform, drawing on experience from across Europe to illuminate policy issues. The European Observatory on Health Systems and Policies is a partnership between the World Health Organization Regional Office for Europe, the Governments of Belgium, Finland, Greece, Norway, Slovenia, Spain and Sweden, the Veneto Region of Italy, the European Investment Bank, the Open Society Institute, the World Bank, the London School of Economics and Political Science and the London School of Hygiene & Tropical Medicine.

Ensuring value for money in health care The role of health technology assessment in the European Union

Corinna Sorenson Michael Drummond Panos Kanavos

Keywords: TECHNOLOGY ASSESSMENT, BIOMEDICAL – organization and administration OUTCOME AND PROCESS ASSESSMENT (HEALTH CARE) EVIDENCE-BASED MEDICINE DECISION MAKING COST-BENEFIT ANALYSIS EUROPEAN UNION

©฀World฀Health฀Organization฀2008,฀on฀behalf฀of฀the฀European฀Observatory฀on฀Health฀ Systems฀and฀Policies All rights reserved. The European Observatory on Health Systems and Policies welcomes requests for permission to reproduce or translate its publications, in part or in full.

Address requests about publications to: Publications, WHO Regional Office for Europe, Scherfigsvej 8 DK-2100 Copenhagen Ø, Denmark Alternatively, complete an online request form for documentation, health information, or for permission to quote or translate, on the Regional Office web site (http://www.euro.who.int/pubrequest).

The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the European Observatory on Health Systems and Policies concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by the European Observatory on Health Systems and Policies in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by the European Observatory on Health Systems and Policies to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either express or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the European Observatory on Health Systems and Policies be liable for damages arising from its use. The views expressed by authors, editors, or expert groups do not necessarily represent the decisions or the stated policy of the European Observatory on Health Systems and Policies.

ISBN 978 92 890 7183 3 Printed in the United Kingdom by MPG Books Ltd, Bodmin, Cornwall.

Contents

List of abbreviations

vi

List of tables, figures and boxes

x

About the authors

xi

Acknowledgements

xii

Executive summary

xiii

1 Project overview, objectives and methods

1

2 Background on innovation and HTA

3

3 HTA and decision-making in Europe

9

4 Conclusions

33

Appendices: Select country case studies Appendix 1: Sweden

39

Appendix 2: The Netherlands

55

Appendix 3: Finland

72

Appendix 4: France

85

Appendix 5: Germany

100

Appendix 6: United Kingdom

114

References

145

List of abbreviations

ACD Afssa Afssaps Afsset ANAES ANDEM ASMR ATC BfArM BMG BZgA CAMTO CBA CBO CDC CE CEA CEDIT CEMTV CEPP CEPS CES CFH CMA CUA

Appraisal consultation document. Agence฀Française฀de฀Sécurité฀Sanitaire฀des฀Aliments. Agence฀Française฀de฀Sécurité฀Sanitaire฀des฀Produits฀de฀Santé. Agence฀Française฀de฀Sécurité฀Sanitaire฀de฀l’Environnement฀et฀ du฀Travail.฀ Agence฀Nationale฀d’Accréditation฀et฀d’Evaluation฀en฀Santé฀ National.฀ Agence฀Nationale฀pour฀le฀Déement฀de฀l’Evaluation฀Médicale. Amélioration฀du฀Service฀Médical฀Rendu.฀ Anatomical therapeutic chemical. Bundesinstitut฀für฀Arzneimittel฀und฀Medizinprodukte. Bundesministerium฀für฀Gesundheit. Bundeszentrale฀für฀gesundheitliche฀Aufklärung.฀ Centre for Assessment of Medical Technology. Cost benefit analysis. Dutch Institute for Healthcare Improvement. Centers for Disease Control and Prevention. Cost-effectiveness. Cost-effectiveness analysis. Comite฀d’Évaluation฀et฀de฀Diffusion฀des฀Innovations฀ Technologiques. Center฀for฀Evaluering฀og฀Medicinsk฀Teknologivurdering (DACEHTA). Commission฀d’Évaluation฀des฀Produits฀et฀Prestations.฀ Comité฀Économique฀des฀Produits฀de฀Santé.฀ College฀des฀Économistes฀de฀la฀Santé.฀ Commissie฀Farmaceutische฀Hulp฀(Pharmaceutical Care Committee). Cost minimization analysis. Cost-utility analysis.

- vi -

List of abbreviations vii

CVZ DAHTA DIMDI ERNIE EUnetHTA FinOHTA HTA HTAi IHE IMOR INAHTA Inserm IOM IQWiG ISPOR ITAS LFN MPA NAM NCC NCCHTA NHG NHI NHS

College฀voor฀Zorgverzekeringen (Health Care Insurance Board). Deutsche฀Agentur฀für฀Health฀Technology฀Assessment.฀ Deutschen฀Institut฀für฀Medizinische฀Dokumentation฀und฀ Information.฀ Evaluation and review of NICE implementation evidence database. European network for Health Technology Assessment. Terveydenhuollon฀menetelmien฀arviointiyksikkö฀(Finnish Office for Health Technology Assessment). Health technology assessment. Health Technology Assessment International. Institutet฀för฀hälso-och฀sjukvårdsekonomi฀(Swedish Institute for Health Economics. Institute for Medical Outcome Research. International Network of Agencies for Health Technology Assessment. Institut฀National฀de฀la฀Santé฀et฀de฀la฀Recherche฀Médicale. Institute of Medicine. Gemeinsamen฀Bundesausschuss/Institut฀für฀Qualitat฀und฀ Wirtschaftlichkeit฀im฀Gesundheitswesen. International Society for Pharmacoeconomics and Outcomes Research. Institut฀für฀Technik-folgen-abschätzung฀und฀Systemanalyse.฀ Läkemedelsförmånsnämnden฀(Pharmaceutical Benefits Board). Läkemedelsverket (Medical Products Agency). Lääkelaitos (National Agency for Medicines). National Collaborating Centre. National Coordinating Centre for Health Technology Assessment. Nederlands฀Huisartsen฀Genootschap (Dutch College of General Practitioners). National health insurance. National Health Service (United Kingdom).

viii Ensuring value for money in health care

NICE NoMA NSF NWO OECD PPB PCT QALY QIS QUORUM R&D RCT RFV ROHTO RVZ SBU ScHARR SHA SHI SHTAC SMC SMR STA STAKES TAB TAG TAR TNO

National Institute for Health and Clinical Excellence. Statens฀legemiddelverk (Norwegian Medicines Agency). National service framework. Nederlandse฀Organisatie฀voor฀Wetenschappelijk฀Onderzoek (Netherlands Organisation for Scientific Research). Organisation for Economic Co-operation and Development. Lääkkeiden฀hintalautakunta (Pharmaceuticals Pricing Board). Primary care trust. Quality-adjusted life year. NHS Quality Improvement Scotland. Improving the quality of reporting of meta-analyses of randomized controlled trials. Research and development. Randomized controlled trial. Riksforsakringsverket (National Social Insurance Board). Lääkehoidon฀kehittämiskeskus (Centre for Pharmacotherapy Development). Raad฀voor฀Volksgezondeid฀&฀Zorg (Council for Public Health and Health Care). Statens฀beredning฀för฀medicinsk฀utvärdering฀(Swedish Council on Technology Assessment in Health Care). School of Health and Related Research, University of Sheffield. Strategic health authority. Statutory health insurance. Southampton Health Technology Assessments Centre. Scottish Medicines Consortium. Therapeutic value. Single technology appraisal. National Research and Development Centre for Welfare and Health. Büro฀für฀Technikfolgen-Abschätzung฀beim฀Deutschen฀ Bundestag. Technical appraisal guidance. Technology assessment report. Kennis฀voor฀zaken (Netherlands Organisation for Applied Scientific Research).

List of abbreviations ix

UNCAM UNOC WHO WTP ZonMw

Union฀Nationale฀des฀Caisses฀d’Assurance฀Maladie. Union฀Nationale฀des฀Organismes฀d’Assurance฀Maladie฀ Complémentaire. World Health Organization. Willingness to pay. Nederlandse฀organisatie฀voor฀gezondheidsonderzoek฀en฀ zorginnovatie฀(Netherlands Organisation for Health Research and Development).

List of tables, figures and boxes

Tables

Table 3.1 Table 3.2 Table A1.1 Table A2.1 Table A3.1 Table A4.1 Table A5.1 Table A6.1 Table A6.2

Institutions and advisory bodies responsible for HTA activities. Criteria for assessment. Overview of HTA governance, processes and role in decision-making in Sweden. Overview of HTA governance, processes and role in decision-making in the Netherlands. Overview of HTA governance, processes and role in decision-making in Finland. Overview of HTA governance, processes and role in decision-making in France. Overview of HTA governance, processes and role in decision-making in Germany. Examples of published and planned NICE guidance. Overview of HTA governance, processes and role in decision-making in the United Kingdom.

10 17 50 67 81 96 110 135 139

Figures

Fig. A1.1 Fig. A2.1

Principal reforms in the Swedish pharmaceutical market. Reimbursement decision process in the Netherlands.

41 57

Methods of disseminating and implementing recommendations. High Health Authority in France. Procedures for a standard HTA at ANAES.

31

Boxes

Box 3.1 Box A4.1 Box A4.2

-x-

86 91

About the authors

is a research officer at LSE Health, London School of Economics and Political Science and a Ph.D. candidate in the Department of Social Policy. Before joining LSE, Ms Sorenson served as a senior policy and planning analyst for the United States’ Food and Drug Administration. Also, she has held various consulting positions in health฀ policy and health economics research. Corinna Sorenson (MPH, MHSA)

is a professor of Health Economics at the University of York, Centre for Health Economics. He has undertaken evaluations over a wide range of medical fields including care of the elderly, neonatal intensive care, immunization programmes, services for people with AIDS, eye health care and pharmaceuticals. In addition, Professor Drummond has authored more than 500 scientific papers and currently serves as PresidentElect of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Michael Drummond (D.Phil.)

Panos Kanavos (Ph.D.) is a senior lecturer in international

health฀policy in the Department of Social Policy and Merck Fellow in Pharmaceutical Economics at LSE Health, London School of Economics and Political Science. He has acted as an adviser to a number of international governmental and nongovernmental organizations including the World Bank, WHO, Organisation for Economic Co-operation and Development, American Association of Retired Persons and ministries of health in over 14 transition and developing countries.

- xi -

Acknowledgements

This project was part of a year-long study: Financing฀sustainable฀health฀care฀in฀ Europe. It was endorsed by Luxembourg’s Ministry of Health, Sitra and the Finnish Innovation Fund. The authors greatly appreciate the helpful comments and suggestions and time taken by the final reviewers – Dr David Taylor and Dr Frans Rutten. We are also grateful to Dr Elias Mossialos for his insightful guidance and contributions to the project.

This study was made possible with the financial support of Pfizer Inc.

- xii -

Executive summary

This report addresses the concepts and controversy surrounding health technology assessment (HTA) in Europe, with a particular focus on selected Member States including Sweden, the Netherlands, Finland, France, Germany and the United Kingdom. Extensive review of these HTA systems produced a number of key findings relevant to a wide range of stakeholders including policy-makers, HTA bodies, manufacturers, health-care professionals and patient organizations. 1.

HTA plays a major part in evidence-based decision-making. Without good evidence, the uptake and diffusion of health technologies is likely to be influenced by a range of social, financial and institutional factors. This may result in suboptimal health outcomes and inefficient use of resources.

2.

Innovation and the actual needs of the health-care system should be linked more closely. Products that provide the most value for investment must be identified and supported and their manufacturers rewarded with appropriate reimbursement and pricing schemes.

3.

Many countries have dedicated HTA bodies, but with somewhat unclear and disparate roles and responsibilities. Groups involved in reimbursement and pricing decisions often differ from those affiliated with independent HTA assessment and clinical guidance development. Divergent processes and roles may hinder the effectiveness and efficacy of the decision-making process and lead to unnecessary duplication of efforts and resource use.

4.

Most review bodies involve a range of stakeholders including physicians, health economists, pharmacists and patient group representatives. Most - xiii -

xiv Ensuring value for money in health care

agencies support some level of involvement from patients and consumers, and a greater role for industry representatives has been proposed. More stakeholder involvement is needed to improve the HTA processes and the implementation of decisions and related policy. This is true of broad HTA networks and partnerships (such as EUnetHTA) that can enhance collaboration between agencies and facilitate innovation in HTA processes and methods. 5.

Some countries consider the evidence and resources required to conduct an assessment, as well as its relevance to the primary clinical and/or policy question. An assessment will not be helpful if there are insufficient data, and may delay access to new treatments. Clinical and policy relevance is especially important as HTAs are useful only if they are expected to contribute to the decision-making process.

6.

Generally, HTA processes lack transparency – from prioritizing decision criteria to stakeholder involvement. More transparency is necessary to ensure an open, systematic and unbiased decision-making process.

7.

There is limited information on HTA’s use in identifying areas of disinvestment. More attention should be paid to identifying topics for potential disinvestment so that ineffective and obsolete products and practices do not remain in the health-care system.

8.

Assessments should adopt a broader definition of value and product benefit by considering patient preferences, quality, equity, efficiency and product acceptability among a wide range of stakeholders. Further exploration to reveal how non-quantifiable factors (e.g. preferences, equity concerns) are accounted for in assessments and decision-making will enable the social implications and constraints of efficient and equitable health to be addressed effectively. In pursuit of this aim, the opinions and experiences of health professionals and individual patients are needed to understand the real-world application and use of a product.

9.

Different countries have diverse technical requirements so it is important that the choice of parameters and methods is substantiated and welldocumented. The model and resulting analysis should be as transparent as possible, and shared with all parties involved in its development.

10. Several technical and methodological hurdles remain. These require further investigation and include the summary measures’ ability to capture other benefits important to patients and the public; generalizability of studies beyond the particular setting or country; inability to account for the opportunity costs of new and expensive technologies; and comparability of measures to ascertain patient preferences. Moreover, as the modelling of cost-effectiveness (CE) becomes more complex, more resources

Executive summary xv

should be devoted to assessing new methods and the resulting impact on uncertainty in decision-making. 11. Assessments should take account of indirect benefits and costs. It would be helpful if review bodies could agree on the inclusion of additional years of life provided by new treatments, as well as opportunity costs related to leisure activities. A better understanding of threshold values, other decision criteria and their application to the overall decision process is needed. 12. The timing of assessments can affect significantly the decision-making process and patient access. Programmes have been introduced to provide more timely information on important products. These programmes should be monitored and evaluated for effectiveness and resulting impact on access to new technologies. 13. In order to maintain the accuracy of assessments and ensure that the most valuable products are on the market, re-evaluation is key to the HTA process. Often the data needed to confirm the cost- and clinicaleffectiveness of a technology can be found only after practical application in the market. This is particularly true for novel products and technologies undergoing fast-track assessment. Systems should be created to allow new clinical and health economic information to be introduced during the assessment process and following market entry. However, there must be safeguards to prevent any re-evaluation leading to inefficiency, resource burden and delayed access to treatments. 14. There is a lack of understanding and evidence on the practical application of HTA from both a process (decision-making) and an outcome (health outcomes, care delivery, costs, research innovation) perspective. More focused research is needed in these areas. 15. The scope of HTA has focused predominantly on pharmaceuticals and, less frequently, on other medical technologies such as devices. There should be further exploration of applying the principles and methods of economic evaluation to preventive measures. Additional research should establish whether (and in what circumstances) such assessments have been conducted, and identify the opportunities and challenges. Overall, HTA can play a valuable role in health care decision-making, but the process must be transparent, timely, relevant, in-depth and usable. Assessments need to use robust methods and be supplemented by other important criteria. Maximization of HTA will enhance potential decision-makers’ ability to implement decisions that capture the benefits of new technologies, overcome uncertainties and recognize the value of innovation, all within the constraints of overall health system resources.

Chapter 1

Project overview, objectives and methods

Almost all Member States have experienced exponential growth in the introduction and uptake of health technologies in recent years – new medicines and diagnostic tools, telemedicine and surgical equipment. Such innovation provides enormous opportunity for governments, providers and patients to realize improved health-care services and outcomes. The rapid diffusion of health technologies has presented governments with unprecedented challenges to provide high quality and innovative care to meet population health needs most effectively while managing health-care budgets and safeguarding the basic principles of equity, access and choice. Consequently, governments increasingly are required to manage scarce resources strategically by investing in those services that deliver the best health outcomes. This equates to care that is affordable, effective, safe and patientcentred. Moreover, innovation is supported adequately, with sufficient market access to new treatments. In recent years, various Member States have developed systems to identify the innovations that provide the best value. The National Institute for Health and Clinical Excellence (NICE) in the United Kingdom was the first national attempt to provide faster access to cost-effective treatments through an evidence-based review process. Review bodies, such as NICE, employ health technology assessment (HTA) to ascertain the relative costs and benefits of health technologies. The resulting evidence is used to support various forms of decision-making, such as reimbursement and pricing. In other words, this information can aid priority-setting for access to limited health-care resources. Beyond ascertaining value, increased use of HTA in this setting signals a desire for a more systematic and transparent process to allocate health-care resources. -1-

2 Ensuring value for money in health care

The operations of NICE and its international counterparts have generated controversy. There are concerns regarding the methods employed during the assessment process; HTA’s role and utility in decision-making and prioritysetting; and the resulting impacts on health care. How are assessments prioritized and who decides? What do authorities mean by evidence? How do HTA methods differ across agencies? What, if any, impact do they have on the assessment results? Is HTA actually employed in a way that aids decision-making? What is HTA’s effect on health care in terms of patient care, innovation and costs? Spearheaded by the London School of Economics, this report aims to address the concepts and controversy surrounding HTA in Europe. The report reviews HTA organizations and processes throughout the European Union (EU) and within selected Member States including extensive case studies on Sweden, the Netherlands, Finland, France, Germany and the United Kingdom, the latter focusing on England and Wales (see Appendices). Broadly, the report is intended to identify and address current considerations regarding HTA methodological and process issues related to the prioritization and financing of modern health care. In particular, it describes the processes and challenges for identifying and prioritizing assessments; assesses and compares current assessment methods and procedures; and highlights the barriers to effective implementation. The report also ascertains the roles and terms of engagement of key stakeholders, and captures the opportunities and challenges for the use of HTA guidance in general priority-setting, decisionmaking and health-care provision. Current literature related to HTA was reviewed systematically, including peerreviewed journals and grey literature sources. Where necessary, reports and other information sources were translated into English. Experts in HTA were consulted in order to supplement the secondary data collection and address any gaps in the evidence available. In conclusion, the review of HTA in Europe and the overarching themes identified in the report should assist in improving the HTA process in Europe and its role in supporting value in health care.

Chapter 2

Background on innovation and HTA

Overview of innovation in health care

Health technology is an indispensable part of any nation’s health-care system. During the past half-century, all Member States and several other countries have increased their technological base for health care – in knowledge and through investments in equipment, devices and pharmaceuticals. As a result, national health-care systems have become increasingly advanced as health-care delivery has introduced a range of technological innovations, such as new medicines and diagnostic tools, telemedicine and surgical equipment. The introduction of new technologies has brought remarkable improvements. Many innovations result in applicable new therapies with significant benefits for patients including improved health, enhanced quality of life and reduced adverse or side effects. Moreover, innovations in clinical practice provide enormous opportunity for physicians and other health-care providers to improve the effectiveness, safety and quality of treatment. On a broader level, technological innovation provides governments with mechanisms to improve the quality and outcomes of national health-care objectives. Many innovations offer significant potential benefits to patients and the healthcare system, but their diffusion can prove problematic in resource-constrained health-care environments. Some innovations produce similar or improved effectiveness and quality of care at significantly lower costs; others increase overall health expenditure (Cutler & McClellan, 2001; Newhouse, 1992). Indeed, the nature and strength of the relationship between health technology and costs are complex and evolving. Moreover, demographic transition (the ageing population) and better-educated health consumers have resulted in -3-

4 Ensuring value for money in health care

increased demand for new medical products and services (Deyo, 2002). This technological imperative frequently is accompanied by expectations of public financing and access that will continue to exert pressure on public budgets in the context of lower economic growth. Governments must strive to attain a balance between innovation, medical progress and productivity gains through more efficient management of health-care systems. New innovations can significantly improve clinical practice, but the rapid growth of medical technology, and the increasing volume of new knowledge from basic and applied clinical research, have made it virtually impossible for care providers to keep pace with treatment advancements. Inappropriate practices and variations in the use of new and existing technologies have encroached into health-care provision across Europe so that the most effective and efficient technologies may not always be employed. Often, inertia and reluctance to change long-standing practices and outdated education restrict the uptake of new, cost-effective interventions. Accordingly, many EU countries face the significant policy challenge of harnessing the benefits of technology and innovation while managing healthcare budgets and meeting public demand and expectations. Countries employ a wide array of approaches to control the costs of health technology and support the optimal use of such products and HTA has assumed an increasing role in national priority-setting and health-policy processes. In recent years, various Member States have developed systems to evaluate innovations – determining their relative value for investment and mechanisms for equitable and accessible treatment provision. In the United Kingdom, NICE was one of the first review bodies established to provide faster access to modern treatments through a systematic review process and to promulgate evidence-based decision-making. HTA: overview and key objectives

HTA originated from growing concern about the expansive diffusion of costly medical equipment in the 1970s and taxpayers and health insurers’ ability and willingness to fund their use (Jonsson & Banta, 1999). Moreover, greater public awareness of health-care rationing decisions and a growing consumerist position on health-care policy required more accountability, transparency and legitimacy in decision-making processes. Decision-makers needed a more comprehensive approach to set priorities and obtain maximum benefit from limited resources, without compromising the ethical and social values underpinning health systems (Hutton et al., 2006). The growth and development of HTA reflected this demand for well-founded information to support decisions on the development, uptake and diffusion of health technologies.

Background on innovation and HTA 5

Since the 1970s, HTA has broadened to encompass a wide range of health technologies including drugs; medical devices; medical and surgical procedures; and the organizational and support systems for care provision (Jonsson, 2002). However, the majority of HTAs have been conducted on pharmaceuticals rather than other technologies such as medical devices and surgical procedures (Hutton et al., 2006). On a broad level, HTA can be defined as: The฀ systematic฀ evaluation฀ of฀ the฀ properties,฀ effects,฀ and/or฀ other฀ impacts฀ of฀ health฀ care฀ technology฀ (International Society of Technology Assessment in Health Care, 2002). More specifically, HTA involves the evaluation of an intervention through the production, synthesis and/or systematic review of a range of scientific and non-scientific evidence.1 The type of evidence considered typically includes the safety, efficacy, cost and cost-effectiveness (CE) of a product. However, HTA is also concerned with the societal, organizational, legal and ethical implications of implementing health technologies or interventions within the health system (Velasco-Garrido & Busse, 2005; INAHTA, 2002). For example, HTA often considers health technologies’ broader macroeconomic impacts on national health-care budgets; resource allocation among different health programmes; regulation; and other policy changes for technological innovation, investment, technology transfer and employment (Goodman, 1998). In addition to ascertaining technologies of value, an effective HTA can reduce or eliminate the use of interventions that are not safe and/or effective, or have insufficient cost-benefits. HTA can also be used to identify existing technologies that may be harmful or ineffective. Less commonly, HTA can also identify underused technologies (e.g. preventive screening, smoking-cessation interventions) and the reasons for this (Asch et al., 2000; McNeil, 2001). For a systematic review of the available evidence on a health technology(s), HTA employs a multidisciplinary framework to address four principal questions (UK National Health Service R&D Health Technology Assessment Programme, 2003). • Is the technology effective? • For whom does the technology work? • What costs are entailed in its use? • How does the technology compare with available treatment alternatives? 1. HTA typically entails 1) identifying the policy question, 2) systematic retrieval of scientific and non-scientific evidence, and analysis, and 3) appraisal of evidence, including judgments regarding the meaning of the evidence. The evidence and its applications then inform the decision-making process.

6 Ensuring value for money in health care

An HTA’s principal aim is to provide a range of stakeholders (typically those involved in the funding, planning, purchasing and investment of health care) with accessible, usable and evidence-based information to guide decisions about the use and diffusion of technology and the efficient allocation of resources. In light of these objectives, HTA has been called “the bridge between evidence and policy-making”, as it provides information for health-care decision-makers at macro-, meso- and micro-levels (Battista & Hodge, 1999). Decision-makers have increasingly relied on the use of HTA to support reimbursement and pricing decisions regarding existing and new pharmaceuticals. HTA also contributes greatly to the knowledge base for improving quality of care, especially by supporting the development (or updating) of clinical practice guidelines and standards for health-service provision (Zentner et al., 2005). Without sufficient, high-quality evidence the uptake and diffusion of technologies are more likely to be influenced by a range of social, financial, professional and institutional factors. This may not produce optimum levels of health outcomes or efficient use of scarce resources. Interface between HTA and innovation

The variety and emerging complexity of health technologies has combined with limited national budgets to produce tensions between delivering costeffective health care and improving or sustaining a country’s manufacturing and research base. The importance of achieving a balance between affordable health care and the use of innovative health technologies has increased. This requires not only consideration of the medical and economic value of a product, but also who benefits from innovations, optimal usage2 and appropriate placement in the spectrum of care (Drummond, 2003). HTA provides important benefits by empowering governments to make valuedriven decisions, supporting innovation and providing patients and physicians with the information for making the best treatment choices. However, HTA’s effectiveness (particularly in encouraging innovation) rests on accurate assessments and the appropriate implementation and use of subsequent recommendations. HTA can encourage innovation if assessments of new technology are performed properly and consider a wide range of 2. Variation in uptake and diffusion can signify the sub-optimal use of technology. Excess use is signified when the costs outweigh the benefits for any additional level of technology diffusion or use. Under-use can occur when the foregone benefits outweigh the costs of additional diffusion or use. Both scenarios are sub-optimal, potentially resulting in economic costs and/or reduced health outcomes.

Background on innovation and HTA 7

associated costs and benefits rather than focusing solely on acquisition costs. In particular, adoption costs need to be measured against the potential broader benefits of integrating the new technology into the health system; budget-driven constraints do not necessarily result in the selection of the most effective or cost-effective products. This may require governments to allow additional funding and flexibility between budgets so that expenditure levels are driven by value rather than arbitrary spending caps (Drummond, 2003). As mentioned, HTA’s value in encouraging innovation and value-added health care also depends on the assessment process – including when and how the review was performed and the resulting decision-making procedures. In particular, the following issues can impact on HTA’s effective use for meeting key objectives (Drummond, 2006; Zentner et al., 2005; Anell, 2004; Busse et al., 2002). • Delays in the HTA process can result in deferred reimbursement decisions, restricting patient access to treatments needed. • Evidence requirements can be a significant hurdle for manufacturers, particularly small, innovative companies. These may discourage pursuit of breakthrough technologies. • As HTA becomes increasingly widespread, assessments are made earlier in the technology diffusion process. This may introduce greater uncertainty into the process and the potential for innovations to appear more, or less, beneficial. • Current assessment methodologies may limit comparability and transferability across countries and studies. • Lack of transparency, accountability and stakeholder involvement in the HTA process can decrease the acceptance and implementation of assessment results. • Low numbers of skilled HTA personnel and limited international collaboration between review agencies can reduce the efficiency and effectiveness of assessments. • Separate processes and organizations for economic assessments, reimbursement/pricing decisions and practice-guideline development may hinder the effectiveness and efficacy of the overall decision-making process through unnecessary use of resources and duplication of efforts. Decision-makers are more likely to utilize HTAs if there are established policy instruments (e.g. reports, practice guidelines) and commitments to use them effectively. Moreover, patient demand and the CE of a technology can change so it is important to review HTA recommendations on a

8 Ensuring value for money in health care

consistent basis. This requires greater participation and collaboration among stakeholders particularly HTA personnel, government officials, the industry, health providers and patients. Without adequate input and understanding of the HTA process, stakeholders may mistrust the evidence and subsequent recommendations. For HTA to be of optimal benefit, the assessment process needs to be linked with innovation and other aspects of policy-making – recognizing the complexities of decision-making that require consideration of subjective and normative concerns. Without these links, HTA may have limited power to inform the policy process and facilitate access to new and effective products. HTA’s role in encouraging innovation and value in health care could be improved by greater understanding of the challenges inherent in the HTA process, as outlined below.

Chapter 3

HTA and decisionmaking in Europe

HTA dates from the late 1970s when the expansion of technology and healthcare costs began to capture the attention of decision-makers (Jonsson, 2002). The introduction and subsequent growth within Europe runs alongside health policies that place greater emphasis on measurement, accountability, value for money and evidence-based policies and practices. Moreover, the advent of randomized control trials (RCTs) and subsequent availability of data; growth in medical research and information technology; and increased decentralization of health system decision-making, all contributed to an increased need for HTA activities (OECD, 2005). In Europe, the first institutions or organizational bodies dedicated to the evaluation of health-care technologies were established in France and Spain in the early 1980s and in Sweden in 1987 (Velasco-Garrido & Busse, 2005; Garcia-Altes et al., 2004). Over the following decade HTA programmes were established in almost all European countries, either in new agencies or institutes or in established academic governmental and non-governmental units (Table 3.1). Broadly speaking, such bodies fall into two categories: (1) independent (arms-length) review bodies that produce and disseminate assessment reports on a breadth of topics, including health technologies and interventions; and (2) entities under government mandates (e.g. from health ministries) with responsibilities for decision-making and priority-setting, typically pertaining to the reimbursement and pricing of heath technologies. The latter serve an advisory or a regulatory function. Many EU countries are supporting these efforts by investing resources in HTA and associated evaluation activities. For example, in 2007 Sweden spent around -9-

10 Ensuring value for money in health care

€5.7 million on its national agency, the Swedish Council on Technology Assessment in Health Care (SBU); and the United Kingdom Department of Health allocated £35 million to NICE (SBU, 2007; United Kingdom House of Commons, 2007). Table 3.1. Institutions and advisory bodies responsible for HTA activities Country Austria

Federation of Austrian Social Insurance Institutions/Drug Evaluation Committee [Hauptverband der österreichischen Sozialversicherungs träger/Heilmittel-Evaluierungs-Kommission]

Belgium

National Institute for Sickness and Invalidity Insurance (INAMI)/ Commission for Reimbursement of Medicines [Institut National d’Assurance Maladie-Invalidité/Commission de Remboursement des Médicaments]

Denmark

Reimbursement Committee/Danish Centre for Evaluation and Health Technology Assessment (CEMTV)

Finland

Pharmaceuticals Pricing Board (PPB)/Finnish Office for Health Technology Assessment (FinOHTA)

France

Economic Committee on Health Products (CEPS)/Transparency Commission [Commission de la Transparence]

Germany

Federal Joint Committee/Institute for Quality and Efficiency in Health Care (IQWiG)/German Agency for Health Technology Assessment (DAHTA).

Italy

Committee on Pharmaceuticals/Italian Medicines Agency (AIFA) [CIP Farmaci/Agenzia Italiana del Farmaco]

The Netherlands

Pharmaceutical Care Committee (CFH) /Health Care Insurance Board (CVZ)

Norway

Pharmaceuticals Pricing Board (PPB)/Norwegian Medicines Agency (NoMA)

Spain

Spanish Agency for Health Technology Assessment (AETS)/Catalan Agency for Health Technology Assessment and Research (CAHTA) [Agència de Evaluación de Tecnologías Sanitarias/ Agència d’Avaluació de Tecnología Mèdica i Reçerca]

Sweden

Pharmaceutical Benefits Board (LFN)/Swedish Council on Technology Assessment in Health Care (SBU)

Switzerland

Federal Office of Public Health (BAG)/Confederal Drug Commission [Bundesamt für Gesundheit /Eidgenössische Arzneimittelkommission]

United Kingdom

NICE/National Coordinating Centre for Health Technology Assessment (NCCHTA) /Scottish Medicines Consortium (SMC)

Source: Velasco-Garrido & Busse, 2005; Zentner et al., 2005.

HTA and decision-making in Europe 11

Additional investment brings growing recognition that the HTA process must be scientifically sound, consistent across applications, transparent and of practical use in both policy-making and health-care practice (Zentner et al., 2005; Jonnson, 2002). Further, more countries are placing greater emphasis on ensuring that the results of HTA are considered in key decisionmaking processes. While European HTA agencies share many of the same basic objectives, their structures have developed separately and currently operate differently across countries. In particular, there are variations in: • responsibility and membership of HTA bodies (governance, decisionmaking, priority-setting); • assessment procedures and methods; • application of HTA evidence to decision-making (criteria and timing of assessments); • HTA dissemination and implementation. Moreover, transparency and accountability are encapsulated in each of these elements of the HTA process. The heterogeneity of HTA activities in EU countries reflects their individual health-care and political environments with differing mandates, funding mechanisms and policy-formulation roles (Velasco-Garrido & Busse, 2005; Banta, 2003). Further, HTA’s use in decisions that influence the diffusion and uptake of technologies can be influenced by a myriad of factors such as income levels, reimbursement mechanisms, regulatory environments and behavioural determinants (e.g. cultural imperative for new technology). As HTA strives to connect policy and evidence, it also reflects the specific needs of decisionmakers within a certain system. Responsibility and membership of HTA entities

Most national HTA bodies can be categorized as serving either an advisory or a regulatory role in the decision-making process, depending on the intent and type of assessment required (Zentner et al., 2005). Advisory bodies, such as those in the Netherlands and Denmark, make reimbursement or pricing recommendations to a national or regional government, ministerial department or self-governing body (Zentner et al., 2005). Regulatory bodies are accountable to health ministries and are responsible for listing and pricing drugs, medical devices and other related services (Zentner et al., 2005). This is the role of HTA agencies in Finland, France, Sweden

12 Ensuring value for money in health care

(LFN) and the United Kingdom. Other groups mainly coordinate HTA assessments and produce and disseminate reports (e.g. Health Council of the Netherlands, Sweden (SBU)). The mandates or responsibilities of the assessment bodies vary according to their general mission and overall policy objectives (Anell, 2004). As one component in the broader health-care decision-making process, HTA programmes typically reflect the current national policy landscape such as the need to contain costs or improve access to a given intervention or service. Economic evaluations often coincide with policies on the reimbursement, pricing and utilization of health technologies (Hutton et al., 2006). Assessments frequently assume a role in providing information to providers through practice guidelines and in supporting decisions on the investment and acquisition of health technology (OECD, 2005). In many countries the health ministry oversees the appraisal process, although independent institutions (e.g. NICE) often are involved in managing various aspects of the assessment (Hutton et al., 2006). In many social insurancefunded health systems, the process is driven predominantly by insurance organizations (Hutton et al., 2006). However, the health ministry provides some degree of oversight even in these countries and, often, the social affairs or security ministry is involved. Evaluation practices also differ. In general, the nature of an assessment determines which organization will conduct the evaluation. Some HTA bodies conduct assessments via in-house committees; others coordinate independent reviews by external bodies such as university research institutions or expert groups (Anell, 2004). Independent reviews present benefits and drawbacks to the assessment process. They may provide greater transparency and help to prevent or resolve disputes (Drummond, 2006; Goodman, 1998). Moreover, decentralization can widen the expertise available and bring a broader range of perspectives to the process. However, independent reviews may introduce certain methodological challenges, such as use of particular study designs (e.g. RCTs) and potential disconnections between the economic model and systematic review. A decentralization of responsibilities may also result in coordination inefficiencies; divergent agendas and methodologies; and opportunities for miscommunication between those conducting the assessment and the ultimate decision-maker(s). HTA entities also have differing roles in the decision-making process when assessments are complete. In some countries (e.g. United Kingdom) the HTA body develops guidance and/or recommends reimbursement status; in other systems this is determined primarily by national authorities, insurance representatives or independent self-governing bodies. Moreover, some HTA

HTA and decision-making in Europe 13

committees (e.g. in Finland and France) are also involved in negotiating product prices and reimbursement with manufacturers. All HTAs have multiple technology-related policy-making needs and perspectives across diverse stakeholder groups. Thus, HTA involves a variety of stakeholders including physicians, pharmacists, health economists, insurance and industry representatives and patients. Anell (2004) found that most reimbursement status recommendations are determined firstly by scientific members (e.g. physicians, epidemiologists) with expertise in evaluation of medicines. These decisions are corroborated by academic entities, representatives from patient organizations, health economists and (in the case of NICE) NHS managers (Anell, 2004). Participation differs across HTA bodies, although all these agencies have some level of stakeholder involvement. A recent OECD study (2005) reported that patients and consumer groups were least involved in the assessment and decision-making process. While patient perspectives were taken into account indirectly through the inclusion of safety, effectiveness and quality-of-life measures, such indicators may not adequately reflect important broader patient values (e.g. preference for one treatment, acceptability of various side effects). Measures of such preferences can play a substantial role in the assessment of new technologies and may provide useful insights into the real-world value of technologies. Greater participation of patients and consumers has been advocated in light of these potential benefits (Coulter, 2004) and some HTA systems support an increased role for patients in assessments and decision-making. NICE has established a Citizens’ Council to gather public perspectives on key issues that inform the development of guidance documents. This assists the development of the social value judgments that should underpin NHS guidance. A greater role for industry representatives has also been suggested. This is controversial as there is concern that greater collaboration between HTA entities and industry might influence the objectivity and transparency of the assessment process, particularly in the use of commercial in-confidence data. As a result, the implementation of recommendations could be hindered by appeals and general disagreement from various stakeholders. Stakeholder involvement is generally resource-intensive, but it can improve relevancy and produce greater trust in the assessment. Accordingly, increased engagement may facilitate better overall assessments, reduce the number of appeals and improve implementation of HTA recommendations and guidance (Drummond, 2006). In particular, by playing a more integral role in the prioritization and assessment of health technologies, patients and their organizations can drive a more consensus-based policy process, especially at the macro-level of the health-care system.

14 Ensuring value for money in health care

Assessment procedures and methods

Assessment processes within the EU differ on a variety of issues such as topic selection, evidence/data requirements, analytical design and the methodological approach(es) employed. Topic selection

Most HTA agencies struggle to keep pace with new technology therefore priority-setting has become an important aspect of the process to determine which products are assessed. Countries set priorities using a number of different mechanisms and criteria, through the emphasis given to different approaches (i.e. proactive, reactive, mixed) and in the process of needs identification. The topic agendas of some review bodies are set by national authorities – typically, the health minister or department of health. However, Germany and the United Kingdom have established processes for suggestions to be submitted by a wide range of stakeholders. In Germany, a board of trustees comprised of public administrators, patients and industry representatives determines HTA topics using a Delphi process (OECD, 2005). Within the United Kingdom’s NHIR Health Technology Assessment Programme, advisory panels recommend priorities to the Director of Research and Development. The Scottish Medicines Consortium (SMC) aims to evaluate every new drug, formulation and indication within 12 weeks of market launch. For review bodies responsible for reimbursement decisions, topics of assessment are based upon manufacturer submissions (a dossier of clinical and health economic evidence to support reimbursement determinations). The breadth of assessment topics varies too – some HTA agencies focus on health technologies (specifically drugs and/or devices); others attend to particular disease areas or health conditions. Several organizations conduct assessments on both products and broader health issues (e.g. SBU in Sweden). The criteria used to select topics vary across agencies, but generally include health benefit(s); impact on other health-related government policies (e.g. reduction in health inequality, improving access); uncertainty about effectiveness/CE; disease burden; potential benefits and impact of the assessment; and innovation capacity (Garcia-Altes et al., 2004; Taylor, 2001; Oortwijn et al., 1999). Generally it is considered not cost effective to evaluate all existing technologies and interventions. Review bodies incorporate various approaches to ensure the efficiency of the assessment process in order to provide important and timely information for decision-makers. In the United Kingdom, NICE allows groups of similar technologies to be compared; the Netherlands requires certain procedures to guide proper use of drugs that are not appraised. If a

HTA and decision-making in Europe 15

drug provides several approved indications then review bodies in Sweden, the Netherlands, Norway and the United Kingdom commonly evaluate the therapy for all conditions. Although HTA agencies cover a quite broad range of topics, some areas are studied less (e.g. low-technology and preventive interventions). This is also true for research on ineffective and obsolete technologies and interventions. Moreover, HTA bodies rarely undertake assessments to keep abreast of new areas of research and development (R&D), presumably because of limited resources (OECD, 2005). Assessments conducted earlier in a product’s life-cycle have had some impact by identifying areas of uncertainty and highlighting areas for further research (OECD, 2005). Similarly, some HTA agencies have developed early-warning and horizon-scanning systems to identify new and emerging technologies that might require urgent evaluation, consideration of clinical and cost impacts, or modification of clinical guidance activities (Douw & Vonderling, 2006; Douw et al., 2003; Carlsson et al., 1998). Criteria used most often to identify candidates for early warning assessments are listed below. • Requires attention of (or action by) politicians, hospitals and health-care administrators within certain time limits. • Deemed cost-demanding or controversial. • Expected to spread more rapidly than desired, according to current scientific knowledge base. • Expected to spread more slowly than desired given technology’s potential benefit. This type of programme has been established nationally in the Netherlands, Sweden, Finland and the United Kingdom; and internationally through the European Information Network on New and Changing Health Technologies (EuroScan).3 There is limited evidence of impacts on decision-making, but there is some concern that early assessment may be biased against new technologies, especially those of higher cost (AdvaMed, 2000). Several studies have highlighted a lack of transparency in the topic selection process (Garcia-Altes et al., 2004; Hagenfeldt et al., 2002; Henshall et al., 1997). Many HTA organizations lack explicit processes for prioritization, including selection methods and stakeholder involvement (Garcia-Altes et al., 2004). It is important to identify the factors involved in the priority-setting process 3. EuroScan has 12 members (predominantly HTA agencies) in 10 countries – 2 outside Europe (Canada and Israel).

16 Ensuring value for money in health care

and the specific objectives as these affect the criteria for selecting assessments. Typically, there is limited mention of any political deliberation (or other normative considerations) that drives the assessment of certain technologies. Given limited resources and greater accountability, it is increasingly important to state how assessment topics are selected. A certain level of transparency is needed for an open, systematic and unbiased assessment prioritization process (Hagenfeldt et al., 2002). Perceived lack of transparency may exacerbate existing tensions about balancing access to technologies, product innovation and health expenditures between manufacturers, patients and the stewards of health-care budgets. Evidence/data requirements

HTA systems require various types and qualities of evidence for economic evaluations (Hutton et al., 2006). Typically, manufacturers are required to submit a comprehensive summary of a product’s effectiveness and CE but these data play different roles in the assessment process. In Austria, Norway and the Netherlands, HTA bodies review and validate industry data, which must be based on systematic review of available clinical and economic evidence (Zentner et al., 2005). Other organizations (e.g. NICE, SBU) perform systematic reviews in-house or through an independent evaluation group. Evidence used in these assessments may or may not include manufacturer data and generally involves broader review of various information sources. Some countries (e.g. France, Switzerland, Finland) prefer, but do not require, systematic reviews. Their assessments are based primarily on a definite number of studies (e.g. pivotal clinical trials) provided by industry (Zentner et al., 2005). Assessment of unpublished evidence (e.g. commercial in-confidence data) is considered explicitly in Austria, the Netherlands, Sweden and the United Kingdom. Manufacturers generally submit evidence comprised of systematic literature searches and analyses of clinical and economic studies, which may or may not include modelling. The majority of HTA institutions have published guidelines to outline the methodological requirements for manufacturers and reviewers. However, such documented procedures require varying levels of detail and transparency (Zentner et al., 2005). It is typical for the pharmacoeconomic methodologies used in assessments to be described in more detail than clinicalreview procedures or the evaluation of other product characteristics (Zentner et al., 2005). Most guidelines cover preferred clinical and economic evidence, comparators, specification of the outcome variable(s), sub-group analyses, costs to be included, time horizon, discounting and use of sensitivity analyses and modelling.

HTA and decision-making in Europe 17

Differences in timing for evidence requirements have developed recently, i.e. the point at which manufacturers submit CE data . The Swedish LFN requires manufacturers to submit evidence on cost-efficacy. If this is acceptable, the product under review is allowed provisional reimbursement while CE data is collected and submitted. Analytical design

Countries employ different analytical frameworks to guide their assessments (Hutton et al., 2006). Most evaluations assess a variety of criteria including safety and clinical effectiveness; patient need and benefit; and CE and cost of therapy (typically in relation to benefit) (Zentner, 2005; OECD, 2005; Anell, 2004). Some HTA bodies also frame the evaluation around other factors, listed and compared in Table 3.2:

Table 3.2. Criteria for assessment

Criteria

AT

4

BE

CH

DE

FI

FR

NL

NO

SE

UK

Therapeutic benefit

X

X

X

X

X

X

X

X

X

X

Patient benefit

X

X

X

X

X

X

X

X

X

X

CE

X

X

X

X

X

X

X

X

X

X

X

X

X

X

Budget impact Pharmaceutical/innovative characteristics

X

Availability of therapeutic alternatives

X

X

X

Public health impact R&D

X

X

Equity considerations

X

X

X

X

X

X X

Source: Adapted from Zentner et al. (2005) and case studies.

4. AT=Austria, BE=Belgium, CH=Switzerland, DE=Denmark, FI=Finland, FR=France, NL=Netherlands, NO=Norway, SE=Sweden, UK=United Kingdom.

18 Ensuring value for money in health care

While the particular analytical framework may depend on the specific policy question, almost all assessments consider therapeutic and patient benefit. There is also agreement that economic evaluations should be conducted from a societal perspective, taking account of costs and benefits outside the health sector rather than a narrow budget perspective on resource use (Zentner et al., 2005; Anell & Svarvar, 2000). Assessment methods5

HTA uses diverse methods, but most programmes employ an integrative approach. The majority of agencies share similar methodologies and emphasize the most rigorous types of studies (e.g. use of RCTs and cost-utility analyses), but there is no standard approach for conducting assessments. Given their varying orientations, resource constraints and other factors, assessment programmes tend to rely on different combinations of methods. In particular, assessments often differ according to the (Zentner et al., 2005):6 • type of economic assessment required • classification of product benefit (benefit vs. harm) – hierarchy of evidence • choice of comparator • specification of the outcome variable • costs included in the analysis • discounting • use of CE threshold • allowance for uncertainty • missing and incomplete data Type of economic assessment

In general, different countries have similar requirements for economic assessments (Zentner et al., 2005). Typically, these are guidelines that product sponsors must follow to select the type of economic assessment used in submissions. However, some countries (e.g. Switzerland) do not require

5. This section refers primarily to decision-making bodies that review clinical and economic evidence for product reimbursement and pricing. 6. Methodologies can also differ on sub-group analyses; time horizons; instruments to measure quality of life; and methods for calculating costs.

HTA and decision-making in Europe 19

assessments and so no guidelines are applied.7 Among existing guidance, CE or cost-utility analyses are most often considered appropriate analytical designs, particularly when the proposed product has significant clinical advantages over the comparator and relative benefits need to be considered against costs. Cost-utility analysis measures health outcomes in terms of quality-adjusted life years (QALYs).8 Increasingly this has become the preferred indicator of effectiveness as it can be applied in comparisons of different therapies and, consequently, employed for priority-setting (Zentner et al., 2005). Moreover, cost-utility analysis is deemed to be associated with fewer issues than other methodological approaches, such as cost-benefit analysis. Although many assessment bodies (such as NICE) have deemed QALYs the principal measure of health outcome, still only a limited number of studies report QALYs based on the actual measurement of patients’ health-related quality of life (HRQoL)9 (Rasanen et al., 2006; Rawlins & Culyer, 2004). Evidence to classify product benefit

Zentner et al. (2005) found that all countries consider head-to-head RCTs with a high degree of internal and external validity to be the most reliable and objective evidence of a product’s relative therapeutic benefit. This also applies to systematic reviews and meta-analyses of RCTs. Moreover, the majority of review bodies favour RCTs in naturalistic settings as they reflect daily routines and country-specific care delivery. Where definitive primary studies exist, limitations must be considered. For example, elderly people and patients with co-morbidities often are excluded from clinical trials even though they are major consumers of medical products. Additionally, trials do not always collect a full range of economic data (e.g. indirect costs, health-utility measures) and the study time horizon is often too short to detect longerterm outcomes. Findings from different types of studies should be combined or synthesized to supplement available clinical data in order to formulate 7. In the case of Switzerland, the assessment body applies a cost-analysis approach, whereby a new product is compared with the same therapeutic category and the price is compared with those in several other EU countries (Denmark, Germany, Holland, United Kingdom, France, Austria, and Italy). 8. QALYs provide a common unit of evaluation across multiple domains, including estimating the overall burden of disease; comparing the relative impact of specific diseases, conditions, and health care interventions; and, making economic comparisons, such as the CE of different health care interventions. 9. HRQoL measures are used increasingly alongside more traditional outcome measures to assess health technologies. These capture dimensions such as: physical, social and cognitive function; anxiety/distress; pain; sleep/rest; energy/fatigue; and perception of general health. HRQoL measures may be disease-specific or general.

20 Ensuring value for money in health care

effective and comprehensive policies. To that end, other types of studies (e.g. case series, registries) may be preferred to RCTs for different policy questions. For instance, modelling is useful when making decisions under uncertainty. When conducting literature searches, selecting studies and assessing the internal and external validity of clinical trials and systematic reviews/meta-analyses, all review bodies apply internationally established standards. These may be guidelines from the Cochrane Collaboration, CONSORT (consolidated standards of reporting trials), QUORUM (improving the quality of reports of meta-analyses of randomized controlled trials) or their own comparable standards (Zentner et al., 2005). Choice of comparator

Assessments are almost always comparative – the product under review is evaluated against some specified standard of performance or other products and treatments. The choice of comparator is significant in determining the outcomes of clinical and pharmacoeconomic analyses so it is crucial to select an appropriate comparative treatment. Moreover, all relevant options must be included adequately to inform decision-making. Zentner et al. (2005) found that HTA bodies typically use two different selection procedures. Some institutions (Finland, Sweden – new pharmaceuticals only) require a product to be compared with up to three well-defined comparators; others (e.g. United Kingdom) require all relevant comparators. The most costeffective existing therapy usually is deemed the most appropriate comparator. However, for practical considerations, HTA bodies often accept an evaluation against routine treatment or the least expensive therapy. Routine treatment for pharmaceuticals is identified by prescription or sales volume; the dosage and delivery of medication must be therapeutically equivalent. Few agencies provide information on how usual practice is determined for nonpharmaceutical treatments. Other review bodies (Switzerland, Sweden) require products to be compared to all therapies of the same therapeutic group, based on the WHO ATC (anatomical therapeutic chemical) classification system.10 This means that only products that are reimbursed or marketed currently can be employed as comparators. An equivalent dosage form should be defined for comparator medications. France combines both approaches by considering drugs in the same therapeutic group; the most frequently prescribed, the least expensive; and the most recently listed (positive list for reimbursement). 10. The ATC classification system is used for the classification of drugs. It is controlled by the WHO and was first published in 1976. Drugs are divided into different groups according to the organ or system upon which they act and/or their therapeutic and chemical characteristics.

HTA and decision-making in Europe 21

The selectors of the assessment comparator vary between countries (OECD, 2003): specified by either the assessment body (e.g. in United Kingdom) or the product sponsor (e.g. in France and Sweden). In the latter case, HTA bodies often require manufacturers and other relevant reviewers to follow specified guidelines as closely as possible when selecting a comparative treatment. Therefore, it is important that manufacturers communicate with review bodies early in the process (i.e. initial phases of study design) (Zentner et al., 2005). In the Netherlands, product sponsors often discuss the comparator a priori with the assessment body, especially in cases considering a more narrow indication than the product’s label for reimbursement approval. In the United Kingdom, Portugal and Switzerland the comparator selection varies between the product sponsor and assessment entity. In most countries the government plays some role in determining the comparator selected. In the United Kingdom, the comparator increasingly is specified by the Department of Health, particularly when it recommends topics to NICE (OECD, 2003). Other review bodies, particularly NICE in the United Kingdom, consider input from stakeholders when scoping study design and comparator selection. Selection of the outcome variable

Assessments tend to use a variety of health and economic outcome measures. As with comparator selection, the outcome measure(s) specification can influence the conclusions of the assessment. Generally, final outcome parameters that reflect long-term treatment objectives (e.g. changes in mortality, morbidity, quality of life) are preferred, but countries have differing selection procedures and specification processes. In the Netherlands, the outcome variable is outlined by the assessment body; in most countries the product sponsors are the key decision-makers for specifying this.11 The choice of outcome variable may depend on the type of analysis to be conducted and intermediate measures are generally accepted if there is a lack of outcome data available (OECD, 2003). However, this type of study requires a strong and scientifically-based association between intermediate effect and final outcome (Zenter et al., 2005). Costs included in the analysis

HTA bodies and governments differ on the type of costs allowed in assessments. The specification of costs typically is related to the purpose of the analysis and

11. Portugal and Switzerland do not directly specify an outcome variable.

22 Ensuring value for money in health care

the overall objectives of the assessment (OECD, 2003). The differences lie in the inclusion of direct and indirect costs. Some countries (e.g. Sweden) allow all costs to be included in assessments; others (e.g. the Netherlands, United Kingdom) use only direct costs.12 There is a lack of agreement on how to account for productivity loss for indirect costs – whether a human capital or friction cost approach13 (Zentner et al., 2005). Moreover, only some (e.g. SBU) include the costs of additional years of life due to a longer lifespan (as a result of treatment). There is also inconsistency on the inclusion of opportunity losses in changes to the quality of life related to leisure activities and time spent on household duties. Some systems (e.g. in the Netherlands) assume a societal-cost perspective despite such costs extending beyond budget constraints. However, wider costs typically are presented separately from system-related costs, therefore non-system costs may have limited direct impact on decisions. Some countries issue guidelines. Zentner et al. (2005) found that HTA entities or governments provide guidelines on the inclusion of costs associated with other diseases resulting from prolonged life and placing utility measures against the preferences of a country’s population. As data are not always transferable across countries, HTA bodies often request resource consumption and related costs based on national data. Moreover, most guidelines require a high degree of transparency in cost calculations. This entails identifying costs accurately, separating the quantity of resources consumed and the respective cost, and detailing adequately any data sources (Zentner et al., 2005; OECD, 2003). Discounting

The use and effects of many products extend for years, especially those for chronic conditions. Where a product impacts health and treatment utilization for over one year it is considered good practice to employ discounting to assess appropriately the changes in costs and benefits over time (OECD, 2003). Almost all HTA bodies employ discounting in assessments, typically applying an annual rate between 2.5% and 10% to both costs and benefits (Zentner et 12. United Kingdom includes only direct costs to the NHS and personal social services. 13. Human capital approach employs the principal that a profit-maximizing firm will employ labour up to a point when the value of the marginal product of labour is equal to the gross wage. Thus, it treats the value of each unit of time lost to ill health to be equal to the gross wage and any additional employment costs. The friction cost approach considers that workers who withdraw from work due to ill health or death will be replaced, after some period of adaptation. Even with very short-term absence, a firm can use existing capacity in its labour pool to compensate for lost work due to ill health.

HTA and decision-making in Europe 23

al., 2005; OECD, 2003).14 However, a discount rate of 5% is recommended to reflect a societal perspective. Either the payer or the product sponsor determines the discounting rate, according to the country. As a general rule, institutions require the discount rate to be included in the sensitivity analysis in order to determine the effects on outcomes. Use of CE threshold

In economic evaluation, the results of a CE analysis are summarized by the CE ratio.15 This compares the incremental cost of an intervention with the corresponding incremental health improvement. The health improvements typically are measured in QALYs gained, so the CE ratio usually is expressed as a cost per QALY gained. Treatments with a relatively lower CE ratio are considered most cost-effective. Essentially, CE ratios indicate which health technologies will provide health improvements most efficiently (Garber, 2000). It can be problematic to interpret the results of CE analyses and, therefore, difficult to decide whether to adopt a particular treatment. As a result, a CE or willingness-to-pay threshold often serves as a general decision rule for ascertaining value for money. An intervention’s CE ratio often is compared to the threshold in order to recommend inclusion or exclusion in the benefits package. Interventions with unfavourable CE ratios may be adopted if other factors (e.g. disease burden, health equity) are a consideration. Few countries employ a formal or fixed threshold, or at least do not make this explicit. For example, NICE maintains that there is no formal threshold, but recent comments by officials and in particular guidances (e.g. on orlistat) indicate a threshold of £20 000 to £30 000/QALY (Devlin & Parkin, 2004). Rawlins and Culyer (2004) suggest that NICE bases decisions primarily on CE ratios below £20 000/QALY. However, as the CE ratio increases there is increased likelihood of rejection on the grounds of CE. Beyond NICE, the available evidence suggests that the threshold for adoption is between US$ 20 000/QALY and US$ 100 000/QALY, with thresholds of US$ 50 000 – US$ 60 000/QALY frequently proposed (Bell et al., 2006).

14. It is recommended that the same rate be applied to both costs and benefits but some countries use different rates (e.g. United Kingdom, Sweden). 15. The intervention under study and its alternative are denoted as 1 and 0 respectively. If C1 and C0 are the net present values of costs that result when the intervention and alternative are used, and E1 and E0 their respective health outcomes, the incremental CE ratio is simply: CE ratio=(C1-C0)/(E1-E0). This ratio, a cost per unit of incremental health effects, is used often as a measure of value.

24 Ensuring value for money in health care

Allowing for uncertainty

Most review bodies either conduct or require sensitivity analyses on all variables that could potentially influence the overall results, or on a subset of inputs (e.g. imprecise estimates only). This is based on the need to test or verify the robustness of assessment findings. Countries have different requirements for sensitivity analyses (e.g. application of uni- or multi-variate methods), so it is important that the choice of parameters and methods employed are substantiated and well-documented. This is especially important when assessing new technologies for which necessary data are seldom evident. Most countries also require some form of modelling to allow for uncertainty in the variables and estimates used in assessments. Typically, models are generated by either manufacturers or the review body, sometimes both. Many review bodies develop models to substantiate the estimates provided in manufacturers’ submissions and to compensate for missing or incomplete data. Increasingly, complex decision analysis models are used to ascertain CE, although these vary in quality. Many lack adequate transparency, thus making it essential to continue independent assessment of models used in economic evaluations. Sensitivity analyses and modelling (as well as sub-group analyses) also may be used to predict the effect of certain patient characteristics (e.g. age, sex, ethnicity) on CE and equity (Michaels, 2006). Guidance from some review bodies (e.g. NICE) suggests that it might be appropriate to provide modelling of the clinical- and cost-effectiveness of treatments for subgroups of patients, but make no explicit recommendations on which variables would be considered ethical. Clear criteria for subgroup analyses, based on specific variables, could provide a framework for incorporating social values into decision-making in an explicit, transparent and consistent way. Missing and incomplete data

Many HTA agencies face analytical challenges when dealing with data from manufacturers or sponsors. Failure to follow specific assessment guidelines may produce data that are incomplete, poorly presented or lack transparency (OECD, 2005). Moreover, sponsors may be asked to produce the same information in various formats for different countries, presumably increasing the costs of compliance and reducing efficiency. The choice of methods can influence significantly the result of an assessment and its comparability across studies and countries. Ultimately they may impact on HTA’s utility in the decision-making process (Boulenger et al., 2005). Unfortunately, there is minimal information on how agencies handle these data issues.

HTA and decision-making in Europe 25

Application of HTA evidence to decision-making: criteria and timing of assessments

Countries employ a variety of HTA evidence to support priority-setting and other modes of decision-making (see HTA dissemination and implementation). In a comparative study by Zentner et al. (2005), all countries16 compared a drug’s therapeutic benefit17 with available treatment alternatives. In fact, this tended to be the leading criterion for assessing a product’s added value in the majority of evaluations. Health-related quality of life is deemed the most appropriate criterion for a technology’s added value from the patient perspective. NICE is one of the few review bodies that has made explicit commitments to include this measure in its assessments and recommendations. As discussed, many decision-makers do not consider CE against a fixed threshold as an absolute decision rule. Other factors are often considered beside efficacy and CE evidence: • necessity (e.g. disease burden and severity) • public health impact • availability of alternative treatments • equity • financial/budget impact • projected product utilization • innovation of product (e.g. pharmacological characteristics, ease of use) • affordability. Rawlins and Culyer (2004) report that NICE usually requires additional justification for CE ratios over £25 000/QALY such as the degree of uncertainty; wider societal costs and benefits; and the particular features of the condition and population using the technology. The Netherlands has an ongoing discussion about adopting a decision framework based on both efficiency and equity criteria. Different thresholds would apply according to disease burden (essentially necessity) with higher CE ratios allowed for the most severe diseases. 16. Austria, Australia, Canada, Switzerland, France, Netherlands, Norway, New Zealand, Sweden and the United Kingdom. 17. A product is considered as having a therapeutic benefit if it demonstrates an improved benefit/ risk profile compared to existing treatment alternatives. In the case of therapeutic equivalence, a drug is typically not accepted for public reimbursement or is subject to a reference pricing system. A therapy with an inferior benefit/risk profile than other viable therapies are not typically reimbursed, even in the case of lower costs.

26 Ensuring value for money in health care

Many countries lack transparency in their decision-making criteria. An analysis by Anell (2004) found that some review bodies rarely, if ever, outline explicitly the relative weight and importance of the criteria used for recommendations. This is especially true of societal-related and non-quantifiable considerations, such as equity and patient’s quality of life. These tend to follow efficacy and CE in their importance in the overall decision process (Zentner et al., 2005). More explicit understanding of both the threshold value and the accompanying criteria and decision rules is important for a transparent and coherent decisionmaking process. It has also been suggested that the CE threshold should be consistent with overall budget constraints and consider equity and fairness as well as efficiency (Rawlins & Culyer, 2004; Towse, 2003). The time required to complete an HTA can affect the application of its evidence. Specifically, there may be a conflict between ensuring comprehensive evaluations and providing timely information to decision-makers. As different countries have divergent approaches to HTA, so the time allocated and required to complete an assessment varies. More specifically, assessments range from two weeks to a few years; although the average is between three months and one year (OECD, 2005). French assessments tend to take less time (e.g. two weeks) than those in the United Kingdom (NICE) and Sweden where one-year assessments are typical (Zentner et al., 2005). The European Commission Transparency Directive (89/105/ EC) requires Member States to make decisions on reimbursement and pricing for new pharmaceuticals to be made within 180 days of marketing authorization (Zentner et al., 2005). Some agencies have addressed the length of time required to complete assessments. Both the SBU and FinOHTA have introduced rapid reviews to facilitate the assessment process and report on emerging technologies. NICE recently instigated single technology appraisals (STAs), a new fast-track procedure to address time concerns regarding standard assessments. STAs place more emphasis on evidence submitted by manufacturers and less on extensive external review. The SMC typically applies an STA approach to its assessments, providing the NHS in Scotland with guidance based on rapid early assessment of the evidence. Variations in the duration of assessments can be attributed to a number of factors. Depending on the overall mission, mandate and policy objectives, some agencies conduct more of an overview, where results can be delivered rapidly. In addition, some countries may have the resources to conduct primary research in situations lacking key data, but this can prolong the assessment (OECD, 2005). The rapid pace of technological development can extend or delay timelines as HTA results become obsolete or require the development of new evaluation methodologies to reflect advances. Skilled

HTA and decision-making in Europe 27

HTA personnel may be unavailable due to resource constraints or the pace of technologies under review. The United Kingdom has striven to address this by offering training fellowships and providing a steady stream of funding for NICE appraisals. This has enabled academic units to build a critical mass of skilled personnel (Drummond, 2006). Early appraisals can have a number of consequences. Generally, less information is available early in a product’s life-cycle and these assessments rely more on manufacturers’ submissions. Early review may also be less able to consider sub-groups and other restrictions, unless they are highlighted in the company submissions. HTA dissemination and implementation

As mentioned, the results or evidence associated with HTA are used on a wide range of decisions to: • plan resource capacities • shape the benefit catalogue • guide treatment provision • inform corporate investment decisions • identify R&D priorities and spending levels • change regulatory and payment policy • acquire or adopt new technology(ies). Almost all countries require assessments to ascertain reimbursement status, although they place differing importance on the economic evidence (Anell, 2004) – France rarely considers such information when determining reimbursement status. Alternatively, some reimbursement committees may require assessments only for patented drugs and new indications, or varying requirements for different types of products such as generic drugs (Anell, 2004). Overall, health economic evidence appears to have the most impact for decisions on drugs with broad use (thus, significant potential budget impact) and when CE varies by indication or patient sub-population. Economic evidence is also used to restrict the use of products, especially innovative and expensive technologies that may not meet firm decision parameters. Reimbursement of these can be confined to certain indications, patient populations, treatment settings and therapeutic positioning (i.e. firstor second-line therapy) (Zentner et al., 2005). In the Netherlands, expensive inpatient drugs that meet certain criteria after an initial assessment (e.g. projected sales higher than 0.5% of total drug sales in the hospital) are granted

28 Ensuring value for money in health care

conditional reimbursement. Additional information on the drug’s real-world CE is gathered during this three-year period. Reimbursement is withdrawn if further evidence does not demonstrate value for money. Conditional approvals can play an important role in minimizing uncertainty by allowing the use of technology under limited conditions. However, their utility is contingent upon gathering additional data and subsequent re-evaluation of the product (OECD, 2005). In general, technologies are reimbursed without conditions when CE and marginal therapeutic and patient benefits against competing alternatives have been established (Anell & Persson, 2005). However, some drugs for severe disease (with a small patient population) or conditions lacking treatment alternatives (e.g. orphan drugs) are covered even if they have poor CE. HTAs also play a role in product pricing and in negotiating special agreements with manufacturers (e.g. price-volume, cash rebates) (Anell, 2004). However, the closeness of these links differs from country to country. Some countries make reimbursement decisions prior to pricing, others (e.g. Sweden, the Netherlands, Finland) consider the reimbursement and price of a product simultaneously before making a final decision. However, the different HTA schemes and cost-containment strategies adopted by countries may not have significant impacts on individual drug prices. Rather, their effect on drug costs may be more indirect, through better definition of the appropriate clinical indications for the use of treatments (Taylor et al., 2004). Assessment results are also used to develop clinical or practice guidelines. Typically, these include recommendations on priority-setting and provide national support to assist decision-makers (e.g. policy-makers, providers) to determine effective models of treatment delivery. However, health economic evidence is currently not used optimally in developing guidelines; a minority of overall recommendations employ guidelines grounded in HTA. Berg et al. (2004) suggest that the lack of integration between HTA evidence and guideline development may be attributable to a number of factors including a disconnect between the requirements of clinical practice and data generated by HTA; the medical profession’s aversion to combining economics and health; and guideline development’s reliance on efficacy and effectiveness data, rather than CE. Consequently, the authors suggest that guidelines are a limited mechanism for influencing the use or uptake of new health technology (Berg et al., 2004). This is likely exacerbated by minimal coordination between guideline-producing bodies and those that set priorities and fund HTA studies, and by limited resources for their implementation. However, guideline development and HTA are beginning to converge in many countries. Despite increasing support for the use of HTA in national priority-setting and health-care policy-making, there remains a paucity of evidence on the

HTA and decision-making in Europe 29

real-world effectiveness of economic evaluation in improving health-care planning, clinical practice, diffusion of technologies or overall health costs. The use of HTA has produced advances in technical and methodological issues, but decision-makers continue to diverge frequently from the principles of economic evaluation (Goddard et al., 2006). In addition, there is relatively weak evidence on the impact of HTA and research development, an explicit link is found in only two countries – the Netherlands and the United Kingdom (Henshall et al., 2002). A wide range of factors may prevent decision-makers from using strict CE criteria to set priorities, or other stakeholders from using HTA products (e.g. reports, practice guidelines) in decisions on health-care provision and innovation. Goddard et al. (2006) argue that methodological shortcomings are not necessarily the main reason for lack of impact. Rather, the wider context of public-sector decision-making places political, institutional and environmental constraints on decisions. While decision-makers may value health economic information (even requesting or requiring its inclusion in the overall evidence base), other aspects of the public policy process result in sporadic and unsystematic application of HTA. On a macro-level, HTA’s orientation in the decision-making process can affect the extent to which evidence is used to inform policy and related prioritysetting. In particular, countries often hold differing views on the use of HTA recommendations (Draborg & Andersen, 2006; Garcia-Altes et al., 2004). Some countries support recommendations on the grounds that experts are best suited to inform decision-making; others prefer decision-makers to be responsible for interpreting evidence and formulating conclusions to reflect political contexts, country-specific or regional conditions, or other normative circumstances (Draborg & Andersen, 2006). However, decision-makers may lack the technical expertise necessary to understand adequately the methodological strengths and weaknesses of a given assessment. Assessment bodies have done much to enhance HTA’s accessibility and usability among different audiences (e.g. policy-makers, health professionals, general public), but improvements are still needed. Although different decision structures provide policy-makers with a wide range of discretion, failure to use available HTA evidence may produce policies that lead to inefficient, ineffective and inequitable health care. Jacob and McGregor (1997) note, “however excellent an HTA may be, if it fails to be used to influence the working of the health-care system, it is without impact and must be considered without value”. The influence of HTA depends on several other considerations, including the information needs of decision-makers; transparency of the economic evaluation

30 Ensuring value for money in health care

and subsequent decision-making; mechanisms for disseminating decisions; and processes for monitoring and reappraising evidence (Hutton et al., 2006; Zentner et al., 2005). The usefulness of recommendations can be limited by incongruities between the societal and long-term perspective of assessments and the short-term horizon of policy-makers (the moving-target problem)18 (Neumann, 2004). The uncertainty inherent in HTA may also hinder its use in decision-making – effective assessments identify areas of under- and overuse, and can have ambiguous effects on price determination (Crookson & Maynard, 2000). Moreover, best use of economic evaluation may be prevented by broader health-system characteristics such as decentralized management; inadequate public resources or “silo” budgeting; and existing incentives for manufacturers and academics to deliver research that is interesting rather than practical and focused (Rutten et al., 2005; OECD, 2003; Cookson & Maynard, 2000). It has been suggested that interest groups exert significant influence on the process. Decision-making may benefit some groups at the expense of others, and particular groups may have sufficient power to affect government choices (Goddard et al., 2006). The Dutch Council for Public Health and Health Care (RVZ) (2006) noted that, thus far: …decisions regarding payment or non-payment for medical treatment are only based on a limited degree on ‘hard’ factors, such as costeffectiveness, and much more on less transparent considerations, as a result of pressure by lobby groups, such as consumer organizations, the media, and so on. This means that limits are indeed being set at present, but on an ad hoc and somewhat random basis. The result is that the available resources are not being deployed as efficiently as possible. However, effective implementation of HTA requires the involvement of key stakeholders such as providers and patients. A recent OECD study found stakeholder acceptance to be one of the key determinants of whether decisions are actually put into practice within health systems (OECD, 2005). It is difficult to assess in practice HTA's ability to maximize health for a given budget. In fact, few countries have formal processes to measure the impact of HTA. The long-term nature of some effects of HTA (e.g. changes in expectations and behaviour patterns of users) and the fact that economic evaluation is just one of many factors influencing policy and practice decisions are obstacles to measuring the impact of assessments (Hailey et al., 1990). 18. Often there is the possibility that by the time an HTA has been conducted, reviewed and disseminated, its findings may be outdated by changes in a technology, how it is employed or its technological alternative for a given problem.

HTA and decision-making in Europe 31

A clear and well-communicated decision-making process must be in place before recommendations can be implemented. Lack of a defined process can create doubts about the legitimacy of decisions and make them less likely to be supported by stakeholders. This may increase the risk of appeal procedures (Drummond, 2006; Neumann, 2004). Furthermore, ill-defined decisionmaking processes may prevent the producers of evidence delivering timely and relevant advice. A clear decision-making process requires identification of an assessment framework that aligns incentives with evidence and health system objectives. Improved transparency and effective dissemination of recommendations also depends on the methods used for implementing decisions (Box 3.1).

Box 3.1. Methods for disseminating and implementing recommendations • Coverage/reimbursement policy • Formulary restrictions • Medical audit/peer review • Clinical guidance • Accreditation • Standards • Media campaigns • Conferences/workshops • Professional education • Web sites and newsletters

HTAs with well-chosen and appropriate policy instruments; a prior commitment to use assessment findings; stakeholder involvement; and realworld applicability (e.g. not too narrowly focused) of the resulting decisions are more likely to influence practice and, ultimately, health outcomes (OECD, 2005; Henshall, 2002). Clinical guidance documents on the use of health technologies are more likely to be adopted when there is strong professional and financial support, in organizations that have established systems for tracking implementation, and when they reflect the appropriate clinical context (Sheldon et al., 2004). Recommendations by HTA agencies and any resulting decisions must be reviewed and re-evaluated regularly to avoid the moving-target problem. This applies to new technologies and to those already on the market. Some countries

32 Ensuring value for money in health care

(e.g. Finland, France, United Kingdom) have a more structured process for reappraisals and conduct re-evaluations at fixed or variable intervals. Others (e.g. Austria, Switzerland) initiate reviews if new characteristics of products emerge (e.g. new or broader indication) or new or better clinical and/or economic evidence becomes available (Zentner et al., 2005). National and international collaboration is one final area that may improve HTA’s impact on decision-making. Improved cooperation between assessment groups can facilitate the development of methodologies; enhance the transferability and transparency of HTA results and recommendations; and potentially improve the efficiency and accountability of the process itself. The variety of HTA activities and multiplicity of customers also necessitate strong collaboration within and between agencies and different entities dealing with HTA. Several countries have increased collaboration by creating standardized assessment guidelines or standards; convening periodic meetings to discuss assessment issues; devising new channels to encourage communication among national HTA groups; and strengthening the role of international assessment organizations (e.g. Health Technology Assessment International – HTAi) operating at the global level (OECD, 2003). Most countries engaged in HTA activities are involved in one or more international organizations. The European Network for HTA (EUnetHTA) was formed recently to connect public national/regional HTA agencies, research institutions and health ministries, in order to enable effective exchange of information and support Member States’ policy decisions. EUnetHTA represents 59 partner organizations, including FinOHTA, IQWiG, DAHTA, the National Coordinating Centre for Health Technology Assessment (NCCHTA) and the SBU. Overall implementation of HTA could be enhanced by ensuring that it adapts to the policy question and needs of decision-makers. Timely, methodologically sound evidence should be available in line with decision priorities, recognizing the various dynamics of health technology markets and the public policy process. Indeed, responsibility for achieving better alignment between assessments and stakeholder needs requires collaboration and effort from users and producers. Increasing engagement of key constituencies (e.g. patients, providers, industry) will make decision-making processes more acceptable, relevant and transparent.

Chapter 4

Conclusions

Without high-quality evidence, the uptake and diffusion of technologies is likely to be influenced by a range of social, financial and institutional factors. This may not produce optimum health outcomes or efficient use of limited resources. HTA is a significant aid to evidence-based decision-making, but it must address the challenges of delivering timely and relevant information that reflects adequately the dynamics of technology and the health-care system in order to provide the information needed for effective decision-making and priority-setting. There is a particular need for greater correspondence between the actual requirements of the health-care system and innovation. Products that provide the most investment value must be identified and supported and their manufacturers rewarded with appropriate reimbursement and pricing schemes. Overall, the benefits of health technologies must be harnessed while simultaneously managing health-care budgets and protecting the basic principles of equity, access and choice. This report identifies several key issues that affect HTA’s usefulness in supporting effective and efficient decision-making and value-added health care. • Many countries have several bodies dedicated to HTA, with somewhat unclear and disparate roles and responsibilities. Lines of division typically separate entities involved in reimbursement and pricing decisions from those engaged in independent HTA assessment and clinical-guideline development. Divergent processes and roles may hinder the effectiveness and efficacy of the decision-making process and lead to unnecessary resource use and duplication of efforts. - 33 -

34 Ensuring value for money in health care

• Most review bodies involve a range of stakeholders including physicians, health economists, pharmacists and patient group representatives. However, patients and consumers – the ultimate end-users of a given technology – have limited roles in most agencies. NICE has sought to enhance their role in assessments and subsequent decision-making by establishing a Citizens Council that allows these stakeholders to comment on priorities and recommendations. A greater role for industry representatives has been proposed; both NICE and the LFN consult with industry throughout the assessment process. Overall, greater stakeholder involvement is needed to improve the implementation of decision and policy and manage uncertainty while simultaneously allowing access to safe technologies. • While the processes for prioritizing assessments differ between countries, the majority of agencies select topics based on health benefit; disease burden; technology relevance and costs; and societal and ethical considerations. Some countries also consider the evidence and resources required to conduct an assessment, as well as relevance to the primary clinical and/or policy question. This is important as HTAs are useful only if they are expected to contribute to the decision-making process. Moreover, if the necessary data and resources are insufficient or lacking, the assessment will not be helpful and may even delay access to new treatments. • There have been improvements in topic selection, but generally the process lacks transparency – from prioritizing decision criteria to stakeholder involvement. A greater level of transparency is necessary to ensure an open, systematic and unbiased decision-making process. • Most agencies focus on assessments of new technologies. More attention to identifying topics for potential disinvestment will ensure that ineffective and inefficient products and practices do not remain in the health-care system. This will help to support real innovation. • Most agencies have published guidelines to steer evidence collection and the review process. The majority of guidelines cover similar requirements (e.g. comparators, costs to include), but some important differences can impact assessments. (1) Countries have different evidence requirements. Some differences are attributable to the particular agency’s overall mission and mandate. For instance, groups involved in reimbursement and pricing decisions tend to rely on manufacturers’ data, which may or may not include systematic reviews of the evidence. (2) Most countries use QALYs as the preferred indicator of effectiveness for their cost-utility analyses. However, only a few studies use QALYs to measure

HTA and decision-making Conclusions in Europe 35

patients’ health-related quality of life, so there may not be sufficient evidence of these benefits. (3) Most countries rely on (and prefer) head-to-head RCTs to demonstrate a product’s relative benefit. Although considered the most objective type of evidence, they have limitations when ascertaining product value. Assessments should not only include observational studies and other important evidence, but also adopt a broader definition of value and product benefit by considering patient preferences, quality, equity, efficiency and product acceptability among a wide range of stakeholders. The opinions and experiences of health professionals and individual patients are needed to understand the real-world application and use of a product. Except those in the United Kingdom, Sweden and the Netherlands, few agencies consider equity issues explicitly in assessments and subsequent decision-making. (4) Assessments should take account of indirect benefits and costs. Several countries include indirect costs in analyses and a broader societal perspective, but there is a general lack of agreement on how to account for productivity losses – whether by friction cost or a human-capital approach. The results of assessments may differ significantly according to the method used. In addition, it would be helpful if review bodies could agree to include additional years of life (due to longer lifespan gained from treatment) as well as opportunity costs related to leisure activities. Evaluations should account for other indirect benefits such as reductions in treatment costs and availability of treatment alternatives in a particular therapeutic market. (5) Few countries apply a fixed or formal CE threshold, although often the evidence suggests a range of thresholds. While the threshold can indicate an organization’s or country’s willingness to pay, other factors are often considered. However, these criteria and accompanying decision rules are rarely explicit. This requires better understanding of threshold values, other decision criteria and their application in the overall decision process. (6) Most countries require sensitivity analyses and/or modelling to allow for uncertainty in the variables and estimates used in assessments. As different countries have different requirements, the choice of parameters and methods must be substantiated and well-documented. This is particularly true when more than one entity is involved in the development and analysis of models. The model and resulting analysis should be as transparent as possible, with collaboration and information exchange between all involved parties. In addition, the validity of evaluations will become more difficult to ascertain as CE modelling becomes more sophisticated. Consequently, more resources should be devoted to assessing new methods of modelling and the resulting impacts on uncertainty in decision-making.

36 Ensuring value for money in health care

• Technical and methodological hurdles remain despite ongoing improvements. These require further investigation and research and include: summary measures’ ability to capture other benefits important to patients and the public; generalizability of studies beyond a particular setting or country; inability to account for the opportunity costs of expensive, new technologies; and comparability between health state elicitation instruments. • The timing of assessments can significantly impact the decision-making process and patient access. There has been a general trend towards setting up new mechanisms for issuing guidance on new technologies prior to, or immediately after, market entry. Several agencies have developed early-warning or horizon-scanning systems to identify new and emerging technologies that might require urgent evaluation. In addition, NICE recently introduced STAs as a fast-tracking tool for assessments. These types of programmes have been introduced to provide more timely information on products deemed of policy, clinical or cost importance. While still relatively new, these programmes should be monitored and evaluated for effectiveness and their resulting impact on access to new technologies. • Assessments are beneficial only if they are employed to support decisionmaking. The involvement of relevant stakeholders facilitates the acceptance and implementation of decisions. Moreover, a transparent and wellcommunicated decision-making process is vital to ensure legitimacy and acceptance of subsequent recommendations. The availability of relevant policy instruments and collaboration between national and international HTA bodies also facilitate effective and efficient implementation of decisions. Initiatives such as EUnetHTA should be supported to enhance the transferability, efficiency and accountability of the HTA process. • Re-evaluation is a key component of the HTA process – maintaining the accuracy of assessments and ensuring that the best products are on the market. It allows for consideration of new data and accounts of uncertainty during the initial valuation process. Often, the data needed to confirm the cost- and clinical-effectiveness of a technology can be truly ascertained only after practical application in the market. This is particularly true of novel products and technologies undergoing early or fast-track assessment. Systems should be created to allow for the introduction of new clinical and health-economic information during the assessment process and following market entry. However, safeguards must be introduced against inefficiency, resource burden and delayed access to treatments. It will be useful to monitor approaches such as that of the LFN in Sweden. This allows products on to the market on a provisional basis while CE data are collected to support manufacturers’ submissions.

HTA and decision-making Conclusions in Europe 37

Some limitations and areas of recommended future inquiry raised by current evidence deserve mention. There is a lack of understanding (and evidence) about HTA’s real-world impact on decision-making processes and (more broadly) health outcomes, care delivery, health-care costs and research innovation. Several challenging questions remain about the circumstances surrounding the practical use of economic evidence in decision-making and priority-setting. Exactly when is it used? How are criteria operationalized and how are they weighted against the broad spectrum of decision factors? For a given disease area or public health problem, has HTA appropriately and accurately identified interventions that have led to improved health outcomes? Has HTA produced better managed health-care budgets or a decrease in healthcare costs? Does HTA provide sufficient incentives to facilitate innovative R&D? Conversely, has this “fourth hurdle” in the reimbursement process deterred manufacturers from investing in new and innovative therapies? How can HTA be applied more broadly? Clearly, these questions need more focused research. Greater efforts should be made to set up a formal evaluation component within the HTA process. The impact of economic evaluation can be enhanced only by securing a better understanding of the decision-making process and the practical application of HTA. There is limited information on HTA’s use in identifying areas of disinvestment. This requires more research to identify ineffective and obsolete technologies and interventions. Assessment methodologies have advanced significantly, but there is limited knowledge of, or publicly available information on, how nonquantifiable factors are considered in the HTA process – particularly equity concerns. It is necessary to ascertain how these issues are addressed, in both assessments and subsequent decision-making, in order to address effectively the social implications and constraints of efficient and equitable health care. There is also a lack of research on the systematic assessment of public health interventions, especially those focused on prevention. HTA to date has focused on pharmaceuticals and, less frequently, other medical technologies such as devices. There should be further exploration of applying the principles and methods of economic evaluation to preventive measures in order to facilitate a more evidence-based approach to important population health issues (e.g. obesity, smoking). Given the limited evidence on the economic evaluation of public health interventions, more research should be funded to identify completed assessments and what they have revealed. Finally, stakeholders in the HTA process play an important yet poorly understood role. Existing evidence shows that stakeholder involvement can lead to greater transparency, relevancy and acceptance of decisions. However, there has been little attention to how they are involved in the assessment

38 Ensuring value for money in health care

process and how and when their perspectives are considered. As one of the guideposts for successful implementation, this requires additional enquiries on the role and influence of various stakeholders, especially patients and consumers. In conclusion, HTA plays a valuable role in health-care decision-making when the process includes transparency, timeliness, relevance and usability. Moreover, assessments must employ robust methods and be supplemented by other important criteria in the decision-making process. Decision-makers who maximize HTA’s potential will improve their ability to implement decisions that capture the benefits of new technologies, overcome uncertainties and recognize the value of innovation within the constraints of overall health system resources.

APPENDICES: select country case studies

Appendix 1. Sweden Overview of health-care and reimbursement systems

The availability of adequate health and medical care is a central tenet of the Swedish welfare state. The 1982 Health and Medical Service Act sets out equal access to health services and good health as cornerstones of the Swedish health-care system19 (Glenngard et al., 2005). Three primary principles underpin the provision of health and medical care in Sweden, in the following order of precedence: 1. human dignity 2. need and solidarity 3. CE. These priorities are reflected in national regulations and law. Health and medical care is considered a public sector responsibility,20 with public ownership and political control organized on three levels – national, regional and local (municipalities). Overall goals and policies are established at national level by the Ministry of Health and Social Affairs and the National

19. According to the Act , “…every county council shall offer good health and medical services to persons living within its boundaries… and promote the health of all residents”. 20. Approximately 80% of health services in Sweden are considered a public sector responsibility. - 39 -

40 Ensuring value for money in health care

Board of Health and Welfare. County councils21 form the basis of the health-care system, responsible for the provision of health care as required by the Health and Medical Service Act. Specifically, county councils plan the development and organization of services according to the needs of their populations and publicly financed health care (Glenngard et al., 2005). Accordingly, councils have a high degree of autonomy and decision-making power for a wide range of activities, including major investments in facilities, new technologies, user fees and private practitioners’ services (Carlsson, 2004). The municipalities (289) are responsible for long-term care of the elderly (i.e. nursing homes and housing) and social services (Glenngard et al., 2005; Carlsson, 2004). Moreover, local government has authority to introduce policies concerning choice of providers, contracting, hospital mergers, new primary-care models and integrated care. The Swedish health system is funded primarily through taxation. Both county councils and municipalities levy proportional income taxes (typically around 30%) on the population. These are used in conjunction with state grants and user fees to cover health-care services (Glenngard et al., 2005; Carlsson et al., 2000). The high tax rate allows for public financing of most health-care services, including the majority of drug costs. Central government is characterized by decentralized power and responsibility for health care. However, it guides the overall direction of the system by ensuring that health care is efficient and in accordance with national objectives and the goals of social welfare policy. Actual responsibility for implementing and administering government policy lies with a number of central administrative bodies. The Medical Products Agency (MPA) oversees the distribution, regulation and financing of pharmaceuticals. It has particular responsibility for regulatory control of drugs and other related products, including providing information about medicines and approving clinical trials and licences (Glenngard et al., 2005; Carlsson et al., 2000). The Swedish national health insurance system finances care under the auspices of the National Social Insurance Board (RFV), which also oversees price negotiations for pharmaceuticals (Carlsson et al., 2000). Historically, a new therapy would be registered and priced by the RFV and the product sponsor following an MPA review of safety and efficacy. Once a price was established, drugs were reimbursed through the health insurance system, typically without evaluation of clinical value or relative CE (Carlsson,

21. There are 21 county councils in Sweden, including 3 large regions (Stockholm, Skane, West), each has over 1.5 million residents (Carlsson, 2004).

HTA and decision-making Appendices in Europe 41

2004). However, the mechanisms related to the distribution, pricing and reimbursement of drugs underwent widespread scrutiny as drug costs escalated during the 1990s. Further, there was a focus on the need for more explicit priority setting, increased transparency for access and quality, and greater opportunity for patients to influence decision-making. In the late 1990s and early 2000s, Sweden passed several reforms in an attempt to curb increasing expenditure on pharmaceutical products (Glenngard et al., 2005) (Fig. A1.1). The New Pharmaceutical Benefits Reform (2002) was passed to increase the cost-effective use of public-financed pharmaceuticals and to ensure equal drug benefits throughout the country. This established a new independent governmental agency to meet this end and to increase the transparency of explicit priority-setting processes (Pharmaceutical Benefits Board, 2002). The introduction of the Pharmaceutical Benefits Board (LFN) produced significant changes in the pricing and reimbursement of drugs in Sweden – decisions are based on CE data rather than reimbursed automatically within the benefit scheme (Anell & Persson, 2005).

Cost-effectiveness

Fig. A1.1. Principal reforms in the Swedish pharmaceutical market.

Pharmaceutical Benefits Board (2002) Formulary committees (1997) Decentralized drug budgets (1998) Changed user-charges (1997) Parallel trade (1997)

Generic substitution (2002)

Cost-control 1997 Source: Adapted from Anell and Persson, 2005.

2002

42 Ensuring value for money in health care

The LFN’s principal aims are to determine if a new drug (or other medical product) should be included in the positive list for public reimbursement (i.e. the Drug Benefit Scheme), and to negotiate with manufacturers to set the price of the product (Zentner et al., 2005). Moreover, the LFN is responsible for reviewing listed drugs22 to ascertain whether they meet certain criteria outlined by the 2002 reform.23 A five-year time frame was set for review of approximately 2000 drugs (Zentner et al., 2005) (see following section: HTA process and procedures). The LFN comprises one director and ten members; announced by the government every five years (Zentner et al., 2005). The Swedish Government appoints four members with special expertise in health economics; the provincial parliament appoints four members with medical expertise. The other two members are representatives of consumer and patient groups. The review process is conducted by a group consisting of certain LFN members (e.g. pharmacists, health economists), as well as two to four external medical experts, typically physicians and nurses. In addition to central governmental structures, country councils have local formulary committees that are responsible for recommendations on the use of pharmaceuticals (Glenngard et al., 2005). HTA governance and organization

Sweden has been at the forefront of HTA within the EU and was one of the first countries to assess health technologies in the early 1970s (Carlsson, 2004). The Swedish Council on Technology Assessment in Health Care (SBU) was established in 1987.24 As the leading HTA entity, the SBU’s primary objective was to improve the efficiency and equity of access to, and use of, technologies proven safe and effective – not cost-containment (Carlsson et al., 2000). As the focal point for HTA activities in Sweden, the SBU’s remit is to provide central government and health-care providers with information on the overall value of medical technologies, especially new therapies, from medical, economic, ethical and social points of view (Glenngard et al.,

22. Drugs with new strengths do not require reviews. 23. Since October 2002, any prescribed drug that qualifies for a subsidy is required to be exchanged for the cheapest comparable generic alternative available at the pharmacy. The MPA determines which drugs are exchangeable. 24. The SBU was formally established as a national agency in 1992, following an independent evaluation required by the central government. This resulted in a significant increase in the SBU budget and demand for systematic reviews and other HTA activities.

HTA and decision-making Appendices in Europe 43

2005; Carlsson, 2004). Specifically, the SBU reviews the benefits, risks and costs of health technologies used in health-care delivery (Glenngard et al., 2005). It also assists in identifying areas requiring further research. The SBU Board comprises representatives from key health-care organizations who set assessment priorities and organize HTA projects. A multidisciplinary team of leading experts from Sweden and abroad is recruited for each assessment project. Further, a number of county councils have formal links with the SBU; a few finance local HTA units. The SBU recently established a formal agreement with the National Board of Health and Welfare and the MPA aimed at improving cooperation within HTA activities in Sweden and the dissemination of guidelines and information. There are several other health assessment bodies at both regional and local levels.25 • Centre฀for฀Assessment฀of฀Medical฀Technology (CAMTO) in Orebro. Established in 1999 with the primary objective of promoting HTA at the local level. Comprises a network of clinicians, experienced practitioners and qualified researchers. External experts often serve as consultants on study design issues and dissemination strategies. Overall, CAMTO conducts primary research, disseminates HTA results locally and proposes new projects to the SBU. • Institute฀for฀Health฀Economics (IHE) in Lund. Established in the mid-1970s to perform economic evaluations and other related policy analyses. Also undertakes independent method development; participates in scientific conferences/meetings; collaborates with external researchers on various health economics projects; and coordinates commissioned courses and seminars. Most IHE projects are funded directly by stakeholders in the health-care sector and most findings are published in scientific periodicals by external publishers and other institutions. • Centre฀for฀Health฀Economics,฀Stockholm฀School฀of฀Economics. Well-respected international HTA body that collaborates on several SBU projects. • Center฀ for฀ Medical฀ Technology฀ Assessment (CMT) in Linköping. Performs assessment studies of medical technology from various perspectives (e.g. social, economic, ethical, medical). Majority of activities commissioned and funded by health-care providers, international research foundations and commercial clients.

25. Although there are other HTA bodies, this case study focuses on the SBU – the leading HTA agency in Sweden.

44 Ensuring value for money in health care

HTA process and procedures

There is not enough time or resources available to review all existing technologies, so the most policy-relevant technologies are prioritized for assessment. The SBU initiates this process by submitting an annual report to the government. This reviews work accomplished, plans for future work and evaluated and projected impacts (Carlsson et al., 2000). In turn, the Ministry of Health notifies the SBU about national objectives and the annual budget. National objectives typically are determined by the Ministry of Health, Swedish Parliament and various health-care organizations, and tend to focus on broad health issues. Individuals, predominantly from the health field, also nominate topics for assessments. Subsequent to topic nomination, the SBU sets priorities for assessment based on a two-fold process. First, an internal filtering process is used to establish all possible assessment options by scanning different fields of interest and devising a list of topics for discussion by project coordinators and the SBU executive committee (Carlsson, 2004). Second, the SBU Board receives a condensed list containing proposals ranked and selected for pilot review. Pilot studies (typically entailing an extensive literature search, the Cochrane database and other sources) ascertain whether there is sufficient existing scientific evidence to warrant a full review. The SBU Board makes a final decision based on the following selection criteria (Carlsson, 2004; Carlsson et al., 2000). • Health impact – topic should have significant impact on health outcomes such as mortality and morbidity. • Breadth of health problem – topic must relate to a common health problem, with significant economic consequences for society. • Societal and ethical considerations – topic may have ethical and social implications; be controversial; or of great concern to the broader public. • Professional or organizational justification – topic’s perceived importance should be demonstrable from an organizational or professional perspective (i.e. technology may have potential significantly to alter clinical practice). • Methodological requirements. • Cost of technology – especially if overall value is in question. • Technological relevance – may be obsolete product but still used extensively. Following a positive final determination the Board appoints a project chairman and an appropriate project team.

HTA and decision-making Appendices in Europe 45

Since the 1990s, the SBU has relied predominantly on systematic reviews as the fundamental assessment methodology rather than performing original research. Each project team, normally 10 members, establishes criteria and conducts comprehensive assessments by systematically searching, selecting, reviewing and evaluating available research findings. Typical criteria are time to follow-up, participant drop-out rates and relevant end points (Carlsson, 2004). Every study that meets the basic criteria is reviewed by at least two members of the project team and classified into one of three quality and relevance levels – low, medium and high. As well as clinical aspects (e.g. preventive, diagnostic, treatment), each assessment contains an economic and, frequently, an ethical and social component. SBU project teams typically employ guidelines or standardized checklists to direct the review process for economic evidence (Carlsson, 2004; Drummond et al., 1996). These outline different evaluation criteria including: study design (e.g. clear relevance and associated hypotheses, analysis perspective); selection of comparator(s) (rational and transparent justification for selection); type of economic analysis (e.g. CE, cost-benefit) and rationale for selected methodology(ies); breadth and quality of effectiveness data; benefit measurement and valuation (appropriate outcome measures); costing (methods of estimation and reporting of quantities and prices); modelling (with clear description and justification, including key input parameters); discounting (time horizon and discount rate provided); allowance for uncertainty (sufficient consideration of uncertainty related to data inputs, extrapolation/modelling, analytical methods); presentation of study results (availability of disaggregated data, information on any incremental and comparative analyses, clear presentation of findings). Once the evidence has been reviewed systematically and results assembled, the draft report is reviewed by selected committee members and then by the SBU Board and a Scientific Advisory Committee. The Board gives final approval in a summary document and list of recommendations. A comprehensive final assessment report, a Yellow Report, is then published and disseminated. HTA findings are monitored and updated as necessary. The scope of assessments can range from expansive – covering broad health problems such as obesity – to more narrow evaluations of single interventions such as Magnetic Resonance Imaging (MRI) (Carlsson, 2004). The wider-ranging reports (characterized by the Yellow Reports described) were characteristic of evaluations in the 1990s and can take several years to complete. This lengthy duration may render the results irrelevant to the needs of policy-makers (Carlsson, 2004). In response to this concern the SBU instituted SBU Alert in 1997. This provides early identification and assessment of new technologies through relevant, policy-oriented information on their potential impact in order to optimize their diffusion (Carlsson, 2004). The

46 Ensuring value for money in health care

Alert promotes communication on important health issues between experts and non-experts. It aims to identify relevant health technologies and assess their relative value and impact on care delivery, and to ascertain areas for additional research. Based within the SBU, the Alert is a joint effort between the SBU, MPA, National Board of Health and Welfare and the Swedish Association of Local Authorities and Regions (SBU, 2006). The Alert programme has a slightly different assessment process for early reviews. New topics are identified via scientific sources; search of the EuroScan database for information from other early warning units; and requests from medical experts and policy-makers (Carlsson, 2004). Potential proposals on new technologies are reviewed by staff and decided by the Board, employing the following selection criteria (Carlsson, 2004): • significant economic consequences • ethical implications • considerable impact on health-care organization • potential for medical breakthrough • concerns significant patient population or affects a common health problem. Typically, an early review assessment involves one external expert and one SBU reviewer. Information is collected and synthesized on the new technology and its associated effectiveness, risks, CE, ethical and social concerns and organizational impact. The SBU collaborates with experts to produce brief assessments that provide timely information to key stakeholders. These Alert Reports are published on the Internet for review and comment and revised accordingly. A network of approximately 4000 health-care professionals receives this information (Carlsson, 2004). SBU also develops special topic papers (White Reports) that explore healthcare problems or interventions that may require assessment (SBU, 2006). These serve as starting points for future systematic literature reviews and are reviewed by project groups and external experts only. Since 2005, SBU has published more than 120 reports, including the following topics: • stroke (1992) • MRI (1992) • prostate cancer screening (1995) • oestrogen treatment (1996)

HTA and decision-making Appendices in Europe 47

• smoking cessation methods (1998) • back pain (2000) • colorectal cancer screening (2002) • obesity (2002) • moderately elevated blood pressure (2004). The LFN establishes appraisal priorities by sales volume in each therapeutic group. The Board considers the three basic principles that underpin the Swedish health-care system for all related decisions for each appraisal. In addition, decisions are based on both the CE and the marginal utility of products (Anell & Persson, 2005). In April and June 2004, the LFN published Working฀ guidelines for the evaluation of already-approved drugs as well as general pharmacoeconomic guidelines. The LFN primarily reviews clinical and economic evaluations submitted by manufacturers as part of their application packages for reimbursement for specific products, rather than particular medical indications (Zentner et al., 2005; Glenngard et al., 2005) (see Table A1.1). However, the Board can make exceptions and authorize reimbursement for a certain indication or patient subgroup. It may allow reimbursement for a more limited indication than a drug’s original MPA licence for market approval. Before any final recommendation, however, manufacturers and the Swedish Parliament may make submissions to the LFN. Any manufacturer dissatisfied with the final decision can appeal to an independent court (Anell & Persson, 2005). Between 2002 and 2005, the LFN reviewed and made decisions on 107 products; the majority approved for unconditional reimbursement (Anell & Persson, 2005).26 HTA dissemination and implementation

The SBU’s findings are disseminated through a variety of channels, depending on the relevant target group(s). These include health-care managers, patients, purchasers, quality-improvement teams, drug-review committees and other decision-makers at regional, county and municipal levels. The delivery mechanisms include the SBU newsletter (over 100 000 copies per issue) and web site; medical and academic journals; and professional conferences, seminars and training sessions. At regional level, the SBU collaborates with the National Board of Health and Welfare, MPA, LFN and a range of professional health-care and insurance organizations to implement the findings 26. There may be other products that were not included on the LFN web site; moreover, sponsors may withdraw submissions before the LFN decision is made.

48 Ensuring value for money in health care

of SBU assessments. Effective dissemination and implementation require local involvement, so the SBU has organized a nationwide network of various experts. These act as local ambassadors to initiate and promote local (frequently regional) efforts to help ensure that decision-makers use reports and that findings are applied in clinical practice (SBU, 2006; Carlsson et al., 2004). The findings of SBU assessments and manufacturer-sponsored economic evaluations (in the case of the LFN) are used to inform decisions and prioritysetting activities, primarily for reimbursement, pricing and clinical policy and practice via the promulgation of guidelines. The LFN typically makes decisions on including or excluding new drugs in the benefit package within a 12-month review process. Since the end of 2003, the LFN has made reimbursement decisions on therapeutic groups including migraine medications, antacids, antihypertensives, asthma medications, antidepressants, cholesterol-lowering medicines, pain relief and anti-inflammatory medications and antidiabetics (Anell & Persson, 2005). Assessments of medications to treat prostate disease, incontinence and gynaecological problems are planned (Zentner et al., 2005). In general, drugs are reimbursed without conditions when CE and marginal benefits have been established from comparisons (Anell & Persson, 2005). However, some drugs with poor CE are covered if the disease is severe (with a small patient population) or there is a lack of treatment alternatives (e.g. orphan drugs). In addition to other central government and assessment bodies, the National Board of Health and Welfare employs assessment results to develop evidence-based guidelines (Glenngard et al., 2005). These are supplied to the government with the overall objective of contributing to the effective use of health-care decisions, within the constructs of health need and an open and transparent priority-setting process. The guidelines include recommendations or decisions on priority-setting and provide national support to assist health-care decision-makers (primarily politicians, civil servants, administrators, providers) in determining effective models of treatment delivery. Specialist associations frequently collaborate in the development of these guidelines and recommendations. Generally, three versions of each guideline are published – for health-care decision-makers, health-care providers and patients respectively. As directed by the government, the Board must report on the guidelines’ projected impact on the practice of medicine (Carlsson, 2004). However, the Board has no direct link with county councils (responsible for regional health-care systems) despite the participation of clinical and economic experts. This may limit the ownership and implementation of guidelines. Since 2006, guidelines have been, or are being, developed for cardiac care, cancer (three most common forms), stroke, venous thrombosis, chronic obstructive pulmonary disease, alcohol and drug abuse, depression and anxiety (Glenngard et al., 2005).

HTA and decision-making Appendices in Europe 49

Sweden’s decentralized health-care system makes it difficult to ascertain HTA’s true impact on decision-making and priority-setting. There is a clear process for the dissemination of results, but it is less clear how such information is used in national and local decision-making. On a national level, there is evidence that certain SBU reports (e.g. stomach pain, smoking cessation) have had an impact on clinical guidance and practice, and facilitated greater support for HTA (Carlsson, 2004). A review of the LFN by Anell & Persson (2005) suggests that health economic evaluation, particularly information on CE, can support decision-making related to reimbursement. However, a minimal percentage of reimbursement decisions to date have been supported by substantial health economic evidence. The majority of LFN decisions have concerned price changes on listed drugs, which normally would not require the support of economic evaluation. Moreover, health economic evidence appears to have the most significant impact on coverage decisions on drugs with broad use (therefore large potential budget impacts) and when CE varies by indication or patient subpopulation. In these cases, the LFN relies more heavily on detailed health economic analyses from manufacturers. It is possible that the LFN will rely increasingly on health economic evidence to support decision-making for reviews of other medicines. At local level, assessments are used most effectively for decisions on intermediate (e.g. hospitals) and clinical resource allocation and treatment guidelines (Carlsson, 2004). County councils have responsibility for drug expenditures and most local formulary committees lack health economic expertise so, although LFN’s decisions influence the recommendations of local formulary committees, local coverage decisions are more restrictive (Anell & Persson, 2005). This results in uncoordinated national and local decision-making. Other factors influence assessments’ impacts on decision-making. These include the time required to complete systematic reviews and evaluate manufacturers’ data (in the case of LFN); policy-makers’ attitudes to economic information; and complex HTA results that do not always provide a clear policy perspective (Carlsson, 2004). Moreover, HTA has a general problem with limited funds and researchers in small countries like Sweden. While this restricts the ability to address the large number of unevaluated technologies, Sweden strives to ameliorate this by strengthening international collaboration on HTA activities. Specifically, the SBU participates in a number of international endeavours including EUR-ASSESS, HTAi, EuroScan and the International Network of Agencies for Health Technology Assessment (INAHTA, EUnetHTA). In addition, the LFN and SBU collaborate on reviews of groups of drugs. In fact, Sweden is a country with significant collaboration between the HTA and reimbursement agencies.

50 Ensuring value for money in health care Table A1.1. Overview of HTA governance, processes and role in decision-making in Sweden

Sweden HTA governance & organization Institutions/committees

LFN – reimbursement and pricing decisions. RFV – pricing decisions. SBU – primary national HTA body. Ministry of Health and Social Affairs/National Board of Health and Welfare – oversees other institutions; Board issues health-care guidelines.

Entities responsible for reviewing HTA evidence for priority-setting and decision-making

LFN.

HTA agenda-setting body(s)

Predominantly Ministry of Health and Social Affairs; Swedish Parliament.

Areas for HTA

New approved and already reimbursed prescription drugs.

Reimbursement requirements and limitations

Reimbursement depends on yes/no decision for inclusion on positive list. In exceptions, conditional coverage given for particular applications or conditions.

Stakeholder involvement

LFN Board – health economists, medical experts and professionals, representatives of consumer and patient groups.

Various health-care decision-makers utilize SBU reports.

SBU – health-care providers, health economists, representatives from health-care organizations. International collaboration

Secretariat of INAHTA located at the SBU; SBU participates in EuroScan, HTA-related trainings and conferences, EUnetHTA, HTAi and WHO’s Health Evidence Network (HEN). SBU also collaborates on multinational projects, most at Nordic and European levels.

HTA and decision-making Appendices in Europe 51

HTA topic selection & analytical design Governance of topic selection

Ministry of Health and Social Affairs, Swedish Parliament, various health-care organizations, health experts and SBU Board.

Criteria for topic selection

SBU: • health impact • breadth of health • societal and ethical considerations • professional or organizational impact • methodological requirements for assessment • technology cost • technology relevance. LFN: Based on manufacturer submission and/or sales volume in each product group.

Criteria for assessment

Therapeutic benefit, patient benefit, CE, availability of therapeutic alternatives, equity considerations.

Criteria outlined or publicly-available

Yes.

Analysis perspective

Societal.

Duration required to conduct assessments

Broad health issues: 3-4 years; single indications or products: shorter period, typically up to 1 year. 27

Evidence requirements & assessment methods Documents required from manufacturer

LFN requires summary of up-to-date scientific knowledge including references, clinical and health economics studies (with modelling, if applicable). Manufacturers must present data on actual prescription volumes. (cont.)

27. Section applies primarily to the LFN.

52 Ensuring value for money in health care Table A1.1. (cont.)

Systematic literature review and synthesis

Yes.

Unpublished data/ grey literature

Yes.

Preferred clinical study type/ evidence

RCT.

Type of economic assessment preferred or required

Cost-benefit-value analysis, cost-benefit analysis, cost-minimizing analysis with constant health status.

Availability of guidelines outlining methodological requirements

Yes.

Choice of comparator

Requires 3 well-defined comparators for new pharmaceuticals, typically – routine practice, non-medical intervention and do-nothing. For positive list approval, product is compared with all drugs in a therapeutic group – oriented on the second and fourth level of the WHO ATC classification.

Specification of outcome variable

Morbidity, mortality, life quality (QALY) and willingness to pay (WTP). Preference for measures under daily conditions or routine treatment.

Sub-group analyses

For sex, age, disease stage or severity, comorbidities, risk factors and treatment strategies (e.g. primary/secondary prevention).

Costs included in analysis

Direct and indirect; pharmaceutical costs established on basis of pharmacy costs.

Incremental analyses required

Yes.

HTA and decision-making Appendices in Europe 53

Time horizon

Period within which main differences of health effects and costs appear.

Equity issues

Equity considered in decision-making. Analysis does not state how this is accounted for.

Discounting

Costs and benefits: 3% (base analysis), 0%, 3%, 5% and 0%, for costs and benefits respectively (sensitivity analysis).

Modelling

Performed by companies and institutions.

Sensitivity analyses

For central assumptions.

CE or willingness-to-pay threshold

No formal threshold, but likely ranges between £25 000-£40 000 employed.

Missing or incomplete data

Reported problems with poorly presented data from sponsors.

Support for methodological development

Not available.

HTA dissemination & implementation Channels for dissemination of HTA results

Yellow, White and Alert reports; SBU newsletter; professional conferences, seminars and courses; academic journals; and guidances.

Use of HTA results

Reimbursement, pricing and health-care delivery (via guidelines).

Evidence considered in decision-making

Severity of condition, evidence of effectiveness, CE, price, equity.

(cont.)

54 Ensuring value for money in health care Table A1.1. (cont.)

Any reported obstacles to effective implementation

Decentralized decision-making structure; policy perspective of HTA results not always clear; attitudes towards economic information in decision-making/priority-setting; time required to complete systematic reviews and evaluate manufacturers’ data (in case of LFN).

Formal processes to measure impact

No formal process, but has participated in EURASSESS project that studies HTA’s effect on coverage of policy decisions.

Processes for re-evaluation or appeals

Following preliminary decision, sponsors’ representatives may present arguments directly to the PPB. Manufacturer can appeal to an independent court if dissatisfied with the final decision.

Accountability for stakeholder input

Stakeholders that subscribe to SBU Alert reports may comment following publication on the Internet.

Transparent/public decision-making process

LFN: Board’s decisions are outlined in a document available on its web site. Includes arguments for each decision.

Sources: Anell & Persson, 2005; Zentner et al., 2005; Carlsson, 2004; OECD, 2003; International Society for Pharmacoeconomics and Outcomes Research (ISPOR), 1999.

HTA and decision-making Appendices in Europe 55

Appendix 2. The Netherlands Overview of health-care and reimbursement systems

Under the Constitution, every Dutch citizen in the Netherlands is entitled to health care founded on social insurance principles. This health-care system is characterized by a complex array of institutions, regulations and responsibilities (Bos, 2000). Public health care, infectious disease control, environmental protection and the regulation of health-care professionals form an integral part of central government, particularly the Ministry of Health, Welfare and Sport (den Exter et al., 2004). Service delivery typically rests with independent practitioners or non-profit service organizations (Stolk & Rutten, 2005; Bos, 2000). Consequently, health care in the Netherlands consists of an interdependent mix of public and private initiatives under the umbrella of central government. On a macro-level, this translates to collaborative and interrelated policy processes and decision-making across public, private and professional stakeholders. The public-private interplay of responsibility and decision-making power extends to the financing of health care. The Sickness Fund Act (1996) implemented a compulsory national health insurance scheme in the Netherlands (den Exter et al., 2004). Originally the sickness fund covered about 63% of the population through the social security system; others were insured through a similar social insurance scheme (for employees of provincial and municipal governmental bodies) or private plans (den Exter et al., 2004). Since January 2006, statutory and private health insurance have been integrated within one comprehensive package – the basisverzekering (basic health-care insurance policy). This covers the entire population under national health insurance and includes all acute care provided by hospitals, general practitioners and specialists; all drug and appliance costs; and transportation. As of 2004, there were 22 sickness funds, all overseen by the Health Care Insurance Board (CVZ) (den Exter et al., 2004). This represents the government, employers, employees, insurance funds, health-care institutions and health professionals. The Ministry of Health, Welfare and Sport implements the Dutch Government’s pharmaceutical policy, guided by the principle of safe and affordable pharmaceutical care for all. The Medicines Evaluation Board (MEB), under the auspices of the MEB Agency, evaluates and regulates access to the market as the quality and appropriate use of pharmaceuticals is integral to public health protection. Specifically, new pharmaceuticals are registered following evidence of quality, safety and efficacy (Stolk & Rutten, 2005). Historically, this registration resulted in practically automatic reimbursement

56 Ensuring value for money in health care

by health insurance bodies (den Exter et al., 2004). However, in attempts to control costs and assure equitable access to pharmaceutical care, the Ministry increasingly has required evidence of CE prior to admission in the benefits package (positive list) covered by the sickness funds. Conversely, the Government can remove ineffective or obsolete pharmaceuticals from the package28 and, under the Pharmaceutical Pricing Act 1996, sets the prices of pharmaceutical products. As indicated, not all registered drugs qualify for reimbursement. In particular, pharmaceuticals are reimbursed by the sickness fund only if they are admitted to either of the schedules in the positive list. Schedule A includes a reference price system29 with groups of substitutive pharmaceuticals; Schedule B lists drugs with alternatives (Fig. A2.1). Pharmaceuticals are included in Schedule A if they are substitutes for existing drugs. In the Netherlands, it is mandated that medicines be merged in one group (a cluster) if they address similar indications and a comparable method of administration, with no clinically relevant differences in their properties (Stolk & Rutten, 2005). It is assumed that at least one medicine in each group is fully reimbursable. If a new pharmaceutical has no available appropriate and mutually replaceable substitute (therefore the reference price system does not apply), the manufacturer can apply for an evaluation from the Ministry of Health, Welfare and Sport30 (Zentner et al., 2005; den Exter et al., 2004). The assessment is conducted by the Pharmaceutical Care Committee (CFH), the body responsible for valuing pharmaceuticals and providing the Ministry with recommendations for the positive list (see the following section HTA process and procedures). Following a CFH recommendation, supported by the Ministry, the pharmaceutical is included in Schedule B for reimbursement.31 Reimbursement conditions for the positive list are uniform for both public and private payers (Stolk & Rutten, 2005). While the majority of products are reimbursed fully, limits can 28. In 1996, pharmaceuticals in the health benefits package were evaluated on the basis of need and effectiveness, in order to streamline and improve the quality of the overall package. The process resulted in the removal of a significant number of pharmaceuticals (as of April 1996). 29. The level of reimbursement is based on the average price of pharmaceuticals in a comparable group. 30. Manufacturers must explain for which part of the positive list the proposed pharmaceutical seems to be qualified. 31. Since 1999, new drugs with higher prices than existing substitutes may be reimbursed if efficacy and effectiveness requirements are satisfied. Moreover, since 2005, a pharmacoeconomic study and budget impact analysis is formally required for a new drug with a premium price.

HTA and decision-making Appendices in Europe 57 Fig. A2.1 Reimbursement decision process in the Netherlands.

Availability of therapeutically interchangeable substitutes?

Yes

Schedule A

No

Favourable pharmacoeconomic profile?

Schedule B

Source: Stolk & Rutten, 2005.

be set. For example, reimbursement may be restricted to a particular patient population or for use by a certain medical specialty. The CFH is part of the CVZ and comprises 18 members, representing pharmacists, specialist physicians, economists, psychologists, epidemiologists and ministry representatives (Zentner et al., 2005). External medical experts support the assessment process according to the particular assessment. HTA governance and organization

The use and visibility of HTA in the Netherlands has grown over the last 20 years. This is primarily a result of expansions in health technology, growth in health-care costs and a subsequent increase in the regulation of medical products (den Exter et al., 2004). In particular, throughout the 1980s, politicians and policy-makers increased the pressure for systematic evaluation of new medical technology to support decision-making and improve healthcare status and provision. The Health Insurance Council32 required all new technological innovations to undergo CE analysis prior to determination of coverage in the benefit package (Berg et al., 2004). Moreover, there was a push to institutionalize HTA and improve coordination of assessment activities throughout the Netherlands. Three influential advisory bodies – the Health Insurance Council, the Health Council of the Netherlands and the National Council for Public 32. Now the Health Care Insurance Board.

58 Ensuring value for money in health care

Health – established a National Fund for Investigative Medicine33 in 1998 (den Exter et al., 2004). Administered by the Dutch Health Research and Development Council (formerly overseen by the Health Insurance Council), the fund’s primary aim was to finance original research in support of scientific excellence and evidence-based policy-making (Berg et al., 2004; Bos, 2000). This included the evaluation of new and existing medical technologies, including the associated CE and social, ethical and regulatory implications for the particular policy question or decision required (Bos, 2000). In essence, this served as the national HTA programme, supported primarily by the Ministry of Health, Welfare and Sport and of Education, Culture and Science (Berg et al., 2004). Recently the fund was replaced by the Netherlands Organisation for Health Research and Development (ZonMw), a merger of the Netherlands Organization for Scientific Research (NWO) and the existing Netherlands Organisation for Health Research and Development (ZON) (see below). Several pivotal reports on streamlining the benefits package and improving the appropriate use of medical products were released following this merger. These further established HTA’s role in the health-care system (Berg et al., 2004; den Exter et al., 2004). More recently, the Dutch Parliament has become increasingly interested in HTA and has requested status reports on these activities from the Minister of Health (Banta, 2003). The Health Council of the Netherlands is one of the primary institutions for economic evaluation. It is funded entirely by the Dutch Government, with budget contributions from various relevant ministries. Established in 1902, this is an independent statutory body that advises the government (ministers and parliament) on medicine, health care, public health and environmental issues (Berg et al., 2004). Standing and ad-hoc committees report on specific topics requested (Health Council of the Netherlands, 2006). Committees comprise council members and external experts (approximately 200) from various medical specialties and scientific disciplines. Typically, these committees evaluate the effectiveness, efficiency, safety and availability of health interventions. Some committees may also examine specific epidemiological and economic aspects, as well as associated ethical, social and legal issues. There are approximately 40 to 50 committees at any given time, each with an average of 10 experts (Health Council of the Netherlands, 2006). The composition of each committee reflects the need for both appropriate scientific expertise and a multidisciplinary perspective. 33. Before the Fund was established, the three Councils advised the Dutch government that HTA should be viewed and utilized as an important aid to decision-making. The Health Council delineated an ideal HTA system and how this could improve the effectiveness of policy decisions.

HTA and decision-making Appendices in Europe 59

The Health Council also serves an alerting function by providing unsolicited advice on various topics and related ministerial policy. Interests and activities range from health prevention and treatment (e.g. cochlear implants for children, bioterrorism, immunization) to nutrition and the environment (e.g. radiation). Work specific to HTA was undertaken during the late 1990s and early 2000s (primarily via an interim Central Committee on Medical Technology Assessment), but such activities were reduced significantly in 2003 due to lack of funding (Health Council of the Netherlands, 2005). However, in mid-2005, the Council published a report recommending that projects specific to HTA be increased and supported. Several other organizations involved in HTA activities in the Netherlands are detailed below (Berg et al., 2004; den Exter et al., 2004; Bos, 2000). • Netherlands฀ Organisation฀ for฀ Health฀ Research฀ and฀ Development฀ (ZonMw). National health council appointed by the Ministry of Health and the NWO to promote quality and innovation in health research and care. Responsible for the programming, priority-setting and allocation of government funds for research projects in health care and prevention. In particular, the ZonMw Health Care Efficiency Research programme has an annual budget of €12.2 million to actively support CE studies and implementation research. The programme focuses on services covered by health insurance including diagnostics; therapy and care; and organizations in all medical and paramedical disciplines. • Netherlands฀ Organization฀ for฀ Scientific฀ Research฀ (NWO). A statutory organization with the primary goal of improving the quality of healthrelated research in the Netherlands. Acts as a national general research council, playing a significant role in the development of science, technology and culture, including the medical sector. Has supported several HTA initiatives and projects over the last 20 years. • Council฀ for฀ Public฀ Health฀ and฀ Health฀ Care฀ (RVZ). Instituted in 1995 as an independent body that advises the government on public health and care. Primarily issues advisory reports on government health-care policy covering prevention, health protection, general health care and care of the elderly and people with disabilities. Moreover, reports cover various policy aspects including insurance, planning, financing, training and patient rights. • National฀Institute฀for฀Public฀Health฀and฀the฀Environment฀(RIVM). Engages in a number of activities related to technology assessment, with the main task of evaluating and monitoring vaccines. Also evaluates certain medical devices, particularly those requiring sterilization.

60 Ensuring value for money in health care

• Netherlands฀ Institute฀ of฀ Primary฀ Health฀ Care.฀ Independent, non-profit research body with broad expertise in health-service research including technology assessment on topics such as quality systems, home-care technologies and the evaluation of professional procedures. Board of Governors comprises health-care providers, health insurers, patients/ consumers and academics. • Netherlands฀ Organization฀ for฀ Applied฀ Scientific฀ Research฀ (TNO). Foremost biomedical technology institute in the Netherlands. Limited assessment activity, but evaluates medical devices and is actively involved in coordinating EU-wide HTA projects (Bos, 2000). Evaluation activities focus on policy aspects of technology development and diffusion, home-care technology and minimally invasive therapies. Established programme on preventive medicine has issued various HTA reports. • Dutch฀ Institute฀ for฀ Healthcare฀ Improvement฀ (CBO). Active in quality assurance and technology assessment. Plays significant role in consensus and guideline development. • Various฀ academic฀ institutes. Erasmus University Rotterdam plays the most prominent role in HTA in the Dutch academic community. The Institute for Medical Technology Assessment (iMTA) is the largest group dedicated to HTA. Its primary focus is economic evaluation of health technologies, as well as quality assessment of health care. Many projects are carried out in collaboration with health-care providers, particularly hospitals, affiliated with the Fund for Investigative Medicine. The Department of Public Health and the Centre for Health Policy and Law are also involved in HTArelated activities. Beyond Erasmus University, the Department of Health Economics at the University of Limburg is involved in HTA activities, indeed virtually all medical faculties and university hospitals in the Netherlands have some involvement in HTA endeavours. While the organizations listed provide much of the driving force, several smaller organizations undertake HTA activities. Consequently, HTA in the Netherlands is neither concentrated on, nor directed by, one national research and policy organization. Unlike in other EU countries (e.g. Sweden) many different entities, with often divergent research agendas and traditions, must come together to support national policy and priority-setting activities. Despite greater coordination and improved integration between entities historically organized around the Fund for Investigative Medicine, better cooperation and harmonization is needed.

HTA and decision-making Appendices in Europe 61

HTA process and procedures

Although HTA has adequate resources they are not sufficient to evaluate all new and existing health technologies. This has resulted in an increased focus on setting priorities in order to exploit HTA’s potential to improve the efficiency and quality of health care in the Netherlands. In the early days of the Fund, most assessments focused on new, high-cost, sophisticated therapies, with minimal concern for existing technologies (Bos, 2000). The submission, selection and funding of projects often lacked direct links to health-care areas of greatest concern or most underdeveloped. Also, few evaluations examined the social, ethical and legal implications of health technology (den Exter et al., 2004). These problems caused a significant disconnect between most HTA research and heath-care need, policy development and decision-making. Concern over these prioritizing methods led to efforts to make the process more explicit and rational, and to incorporate social and scientific criteria to determine HTA priorities (Berg et al., 2004). Throughout the 1990s, the primary actors in the identification and setting of HTA priorities (e.g. CVZ, Health Council, Council for Medical and Health Research – (ZonMw) undertook formal processes to identify technologies or areas of health care requiring assessment34 (Oortwijin et al., 2002). The technologies were ranked according to a range of criteria: degree of uncertainty concerning efficacy and effectiveness; frequency of use; costs; impact on morbidity, mortality and quality of life; and rate of use (Berg et al., 2004). Priority topics for evaluation included ultrasound therapy, treatment for urinary incontinence, long-term psychotherapy and diagnostic testing. Currently, research proposals are submitted to the CVZ for evaluation and reviewed by members of the Committee for Investigative Medicine and the Board. Specifically, the reviews independently evaluate, rate and score the policy relevance of the submitted proposals based on a variety of objective criteria35 (Oortwijn et al., 2002): • actual burden of disease, given current treatment strategies for the individual patient; • potential benefit for the individual patient;

34. Typical formal process – expert consultation and topic nomination; prioritization via Delphibased process; further ranking; creation of nominated HTA-subject lists. 35. Researchers submitting proposals are explicitly requested to provide information about the policy relevance of the proposed project. Moreover, reviewers apply a standardized weighting system to the selection criteria to rate and score each proposal.

62 Ensuring value for money in health care

• number of patients; • intervention’s direct costs per patient; • financial consequences of applying the intervention over time (impact on total costs of health care); • additional aspects with impacts on health policy (e.g. potential rate of diffusion). Proposals considered to have intermediate to high policy relevance are sent to the CMHR (Oortwijn et al., 2002). These are appraised for scientific quality and accepted, declined or recommended for revision before resubmission. The Health Council of the Netherlands prioritizes assessments according to requests from ministries and parliament and of its own volition. Topics for further assessment and funding are generated directly by the Ministry or via input from expert working groups that formulate funding programmes. The various priority lists described above also help to shape these agendas. In addition to the publication of advisory reports requested by government, the Health Council’s remit includes horizon-scanning. This draws attention proactively to health issues and developments that may be relevant to government policy and associated agenda-setting (Health Council of the Netherlands, 2005). The Health Council’s primary scanning activities focus on preventive and curative health care; nutrition and food quality; environment and health; and work and health. Its secretariat participates in EuroScan to identify significant emerging health technologies for preventive and curative health care. Horizon-scanning also involves the identification of ethical and legal aspects of public health-related scientific developments that may have policy implications. Various assessment and research organizations in the Netherlands complete numerous HTA assessments and research each year, based primarily on extensive literature review (e.g. systematic literature review, meta-analysis) and consultation with expert groups (Bos, 2000). In general, the Health Council and CVZ publish some 20 to 30 reports annually. The range of assessments is quite expansive, as illustrated below (Health Council of the Netherlands, 2005; Bos, 2000). • use of biosynthetic human growth hormone treatment (CVZ); • use of lung transplantation (CVZ); • use of diagnostic imaging techniques for back pain (CVZ); • extra-corporeal membrane oxygenation treatment in neonates (Health฀ Council);

HTA and decision-making Appendices in Europe 63

• cholesterol-lowering therapy (Health฀Council); • silicone breast implants (Health฀Council); • nanotechnologies (Health฀Council); and • use of antiviral agents and other measures in an influenza pandemic (Health฀ Council). As described, the CFH allows reimbursement for new drugs that cannot be substituted, but only if efficiency and effectiveness requirements are met. Manufacturers provide evidence to support the valuation process, including systematic literature reviews or meta-analyses, clinical studies and pharmacoeconomic evaluations (with modelling, if appropriate); consensus guidelines and prescription data (Zentner et al., 2005). In addition, manufacturers typically choose the comparator in accordance with pharmacoeconomic guidelines and may discuss the selection a priori with the relevant assessment body (see Table A2.1). The CFH assesses new medications by comparison with the relevant positive list on a range of criteria, including (Zentner et al., 2005): • therapeutic value • patient benefit • CE • financial impact on benefits package, pharmaceutical and health budgets, Sickness Fund and Dutch society. Therapeutic value and patient benefit is determined by comparison with standard or usual therapies. The CFH employs several legally regulated criteria to evaluate the relative therapeutic benefits, particularly efficacy, effectiveness and potential use of a product. Product use is categorized by three classifications (measured by number of prescriptions over time) (Zentner et al., 2005): satisfactory, broad and limited. Satisfactory or broad use are preferred but limited application does not necessarily result in a negative valuation, especially in cases where the comparator is more expensive. Other relevant criteria include the availability of therapeutic alternatives; disease severity; target patient population; the mode, frequency and comfort of drug delivery; and impact on the quality of life (Zentner et al., 2005). All criteria are important but efficacy, effectiveness and the side-effect profile carry greater weight. Whether the product is a breakthrough therapy or the only available treatment for a condition(s) are critical factors in determining additional therapeutic benefit (Zentner et al., 2005). Other criteria, such as affordability and leakage (use of a product outside the designated patient group) are also considered , albeit less often, in reimbursement decisions (Stolk & Rutten, 2005).

64 Ensuring value for money in health care

Having assessed clinical value, the CFH requires economic evaluations (e.g. CE or cost-benefit analysis) of pharmaceuticals for which manufacturers claim therapeutic benefits. Moreover, since 2005, new drugs with a price premium have been required to undergo economic studies and budget-impact analyses (Stolk & Rutten, 2005). These must be carried out and submitted by manufacturers. In 1999, the Health Insurance Board issued pharmacoeconomic guidelines to standardize this research across the Netherlands and, particularly, manufacturers’ applications for inclusion in the benefits package (Zentner et al., 2005). Two pharmaceutical categories are exempt from the standard valuation process, as previously described. The Ministry of Health, Welfare and Sport has decreed that orphan drugs (treatments for conditions of low prevalence, typically